Options

(EIGRP experts?) Does Static EIGRP neighbors fix hold time expired issues?

itdaddyitdaddy Member Posts: 2,089 ■■■■□□□□□□
%DUAL-5-NBRCHANGE: IP-EIGRP(0) 1: Neighbor 10.1.1.1 (Ethernet0/0) is down:
holding time expired

on my gre/ipsec tunnels i get this a lot and i dont want my eigrp neighbros to drop.
if i set them both to static p2p neighbors will that fix my up and down issues
and is there anything bad that can come from this?

I dont want the eigrp to drop or need to discover a new adjacency

Comments

  • Options
    networker050184networker050184 Mod Posts: 11,962 Mod
    You need to figure out why the timers are expiring and fix that. Don't fix the symptom, fix the problem.
    An expert is a man who has made all the mistakes which can be made.
  • Options
    itdaddyitdaddy Member Posts: 2,089 ■■■■□□□□□□
    headend packet loss i am sure ;( haahah sucks
  • Options
    Harry RolesHarry Roles Member Posts: 19 ■□□□□□□□□□
    Setting static neighbors changes the underlying transport from multicast to unicast. It has nothing to do with neighbor monitoring.
  • Options
    negru_tudornegru_tudor Member Posts: 473 ■■■□□□□□□□
    Is see you have mention of Ethernet (not FastEthernet) so I'll assume this is GNS3 with IOU. It might be worth forcing full-duplex on all your interfaces. I recall from SWITCH studies that I ran into issues because the Ethernet ports were working in half-duplex mode. If it's GNS3 do you have too many routers in that topology? Could be that they're tanking the GNS3 VM resources (might be worth trying IOS for 3725 or 3745 if it's just EIGRP for learning/ testing?).
    2017-2018 goals:
    [X] CIPTV2 300-075
    [ ] SIP School SSCA
    [X] CCNP Switch 300-115 [X] CCNP Route 300-101 [X] CCNP Tshoot 300-135
    [ ] LPIC1-101 [ ] LPIC1-102 (wishful thinking)
  • Options
    shortstop20shortstop20 Member Posts: 161 ■■■□□□□□□□
    You need to figure out why the timers are expiring and fix that. Don't fix the symptom, fix the problem.

    Agreed.
    CCNA Security - 6/11/2018
    CCNP TShoot - 3/7/2018
    CCNP Route - 1/31/2018
    CCNP Switch - 12/10/2015
    CCNA R/S - 1/14/2015
  • Options
    Legacy UserLegacy User Unregistered / Not Logged In Posts: 0 ■□□□□□□□□□
    itdaddy wrote: »
    %DUAL-5-NBRCHANGE: IP-EIGRP(0) 1: Neighbor 10.1.1.1 (Ethernet0/0) is down:
    holding time expired

    on my gre/ipsec tunnels i get this a lot and i dont want my eigrp neighbros to drop.
    if i set them both to static p2p neighbors will that fix my up and down issues
    and is there anything bad that can come from this?

    I dont want the eigrp to drop or need to discover a new adjacency

    When the eigrp goes down on the gre tunnel are the tunnels going down as well?
  • Options
    itdaddyitdaddy Member Posts: 2,089 ■■■■□□□□□□
    reason you get this kind of errors I feel is ESP packets getting crushed or FIFO is effect preventing ESP packets from stream corretly. or what i just say is IPSEC is being crushed by dominated traffic types.

    Solved: you guys are right.
    I adjusted keep a lives first and that stopped the tunnel down a lot.
    keepalive 10 3 change to

    keepalive 10 6 - more retries and that solved the tunnel dropping or down state.

    ----I know I am applying the band aid and not fixing it---agreed-- jut buying some time

    I put in place QOS band policy-map to alot 30 Mbps for ESp traffic and on the other end it is not done. I am trying to get the telco to place QOS commands on my other end point where Internet and ESP traffic converge. It is there where they fight for FIFO traffic and I need QOS prioritizing ESP over Interent. The logical vpn esp traffic rides over a Internet circuit so I want priority for ESP over Internet or really 30 % of a 100 Mbps pipe for esp traffic. I am in the process of getting this done. I solved it cause I forgot I placed QOS policy on my end of the switched cloud but not on the other end. Was studying QOS and dawned on me it IS NOT END TO END QOS. that is my problems. The ESP stream is being joked out at the other end that has not QOS at it termination point. Problem solved..

    logs below. Most of the time 99% of time eigrp timed out but tunnels stayed up. once in a blue moon the tunnel would drop down state.
    but most of the time EIGRP hold time expired. pipe choking at the non qos end point. SOLVED!
    *Oct 23 15:09:32.321 CDT: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.192.1 (Tunnel0) is down: holding time expired
    *Oct 23 15:09:36.157 CDT: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.192.1 (Tunnel0) is up: new adjacency
    *Oct 23 15:10:06.301 CDT: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.192.1 (Tunnel0) is down: holding time expired
    *Oct 23 15:10:08.321 CDT: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.192.1 (Tunnel0) is up: new adjacency
    *Oct 23 15:10:37.529 CDT: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.192.1 (Tunnel0) is down: holding time expired
    *Oct 23 15:10:56.649 CDT: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.192.1 (Tunnel0) is up: new adjacency
    *Oct 23 15:11:11.649 CDT: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.192.1 (Tunnel0) is down: holding time expired
    *Oct 23 15:11:20.989 CDT: %LINEPROTO-5-UPDOWN: Line protocol on Interface Tunnel0, changed state to down
    *Oct 23 15:11:30.993 CDT: %LINEPROTO-5-UPDOWN: Line protocol on Interface Tunnel0, changed state to up
    *Oct 23 15:12:39.037 CDT: %DUAL-5-NBRCHANGE: EIGRP-IPv4 100: Neighbor 192.168.192.1 (Tunnel0) is up: new adjacency



Sign In or Register to comment.