Many thanks to Tony and Ray for weighing in.

I believe I know how the hello-interval and hold-time are supposed to work
according to the documentation. What I experienced in the lab is behavior that
was unexpected and that I could not explain.

After experimenting with this a bit, here is what I have observed.
•       If you configure the physical Serial, Frame-Relay interface with “ip
hello-interval eigrp 1001 20” it will begin working with a hello-interval of
20 seconds and a hold-time of 180 seconds. If the configuration is saved and
the router is reloaded, the configuration on the physical Serial, frame-relay
interface will change with the additional command “ip hold-time eigrp 1001
15” thus telling the EIGRP neighbor to use a hold-time of 15 seconds. The
result is a continuously flapping EIGRP neighbor relationship.
•       If you attempt to apply the command “ip hold-time eigrp 1001 180” to
the physical Serial, frame-relay interface, the command will appear to be
accepted, but no actual changes to the EIGRP neighbors will occur. The command
does not appear in the interface configuration of the Serial interface even
after a save and reload.
•       If you configure either a multipoint Serial, Frame-Relay interface or a
point-to-point Serial, Frame-Relay interface with “ip hello-interval eigrp
1001 20” it will begin working with a hello-interval of 20 seconds and a
hold-time of 180 seconds. If the configuration is saved and the router is
reloaded, the configuration on the multipoint or point-to-point Serial,
frame-relay interface will NOT change. If you add the additional command “ip
hold-time eigrp 1001 140” thus telling the EIGRP neighbor to use a hold-time
of 140 seconds the configuration does begin working as expected even after a
reload

So in conclusion, the issue seems to be related to configuring a Physical
Serial, frame-relay interface to use a hello-interval of 20 seconds. The
configuration will initially work fine, but if the configuration is saved and
the router is reloaded, the router using the Physical Serial, frame-relay
interface will change it’s configuration such that the neighbor router will
be instructed to use a hold-timer of 15 seconds. I was only able to fix this
condition by removing the command “ip hold-time eigrp 1001 15” from the
Serial interface configuration.

Unless I come across new information on this issue, it is time to move on and
concentrate on other topics.

Kind regards, 
jpn
------ Original Message ------
Received: 08:03 PM EDT, 07/06/2013
From: Tony Singh <[email protected]>
To: rayarago <[email protected]>Cc: Jim Newell <[email protected]>,
"<[email protected]>" <[email protected]>
Subject: Re: [OSL | CCIE_RS] Wkb1, Lab 9, Task 9.9

Command Default
The hello interval for low-speed, nonbroadcast multiaccess (NBMA) networks is
60 seconds and 5 seconds for all other networks.
Command Default
The EIGRP hold time is 180 seconds for low-speed, nonbroadcast multiaccess
(NBMA) networks and 15 seconds for all other networks.

http://www.cisco.com/en/US/docs/ios-xml/ios/iproute_eigrp/command/ire-i1.html#wp3171463312


--
BR

Tony

Sent from my iPad



On 6 Jul 2013, at 14:52, rayarago <[email protected]> wrote:

> 
> There are a couple of concepts at play here:
> 
> 
> 
> 1) Default values for EIGRP timers: hello and hold
> 2) What happens when you have default values and you reload?
> 3) The relationship between Hello and Hold in EIGRP
> 4) Neighbors and EIGRP how is it different from OSPF?
> 
> 
> Answers/tips:
> 
> 1) the default values have a loose relationship but they are technically
independent, setting the Hello Timer does not automatically change the Hold
timer (as in OSPF).
> 2) This is what you are seeing, it is not gremlins or bugs: this is what the
router is supposed to do. Sometimes default commands do not make it into the
config on reload it depends on what version of code and what features you are
using.  In the end the point is that you shouldn't think it is a bug.
> 3) review the documentation, VoDs etc.
> 4) with EIGRP you tell your neighbor the schedule for your Hellos, and the
Hold time. IN OSPF 
> 
> 
> BTW, for T1 speeds and below EIGRP has one set of timers and for above it
has another.  It doesn't have anything to do with NBMA, as far as I know. 
Look it up Jim, and let us all know what you found.  (please)
> 
> The only NBMA rule that i can recall is that Split Horizon is disabled for
NBMA.
> 
> Regards,
> 
> Ray
> 
> 
> 
> 
> On Jul 6, 2013, at 3:21 AM, Tony Singh <[email protected]> wrote:
> 
>> 
>> 
>> Make your config more uniform
>> 
>> 
>> ip hello-interval eigrp 1001 20
>> ip hold-time eigrp 1001 60
>> 
>> 
>> Need to change the hold time as the default over NBMA networks is 180
seconds
>> 
>> 
>> On your R5 config you have
>> 
>>> ip hold-time eigrp 1001 15
>> 
>> 
>> So when a hello comes in at 20 secs R5 resets the session as it has not
received it within a 15 sec holdown
>> 
>> --
>> BR
>> 
>> Tony
>> 
>> Sent from my iPad
>> 
>> On 5 Jul 2013, at 22:38, "Jim Newell" <[email protected]> wrote:
>> 
>>> This is in regard to Workbook 1, Lab 9, Task 9.9.  I am not sure if anyone
has
>>> run into this issue before.
>>> 
>>> The same command is applied to the frame-relay interfaces on R2, R5 and R6
to
>>> set the EIGRP hello interval:
>>> 
>>> ip hello-interval eigrp 1001 20
>>> 
>>> Afterward, all three routers are using a hold timer of 180 seconds for
the
>>> EIGRP neighbor on the frame cloud and the neighbors are fully up.
>>> 
>>> R2#sh ip ei nei
>>> IP-EIGRP neighbors for process 1001
>>> H   Address                 Interface       Hold Uptime   SRTT   RTO  Q 
Seq
>>>                                          (sec)         (ms)       Cnt Num
>>> 0   150.100.100.5           Se0/1/0.256      167 00:01:47 1251  5000  0 
557
>>> 
>>> R5#sh ip ei nei
>>> IP-EIGRP neighbors for process 1001
>>> H   Address                 Interface       Hold Uptime   SRTT   RTO  Q 
Seq
>>>                                          (sec)         (ms)       Cnt Num
>>> 0   150.100.100.2           Se0/1/0          166 00:02:08  103  2646  0 
124
>>> 
>>> If I save the configuration on either R5 or R6 and reload the router, the
>>> running configuration that appears on the serial interface changes. This
>>> behavior is exactly the same on R5 and R6.
>>> R5 before reload:
>>> 
>>> interface Serial0/1/0
>>> bandwidth 64
>>> ip address 150.100.100.5 255.255.255.0
>>> ip bandwidth-percent eigrp 1001 43
>>> ip hello-interval eigrp 1001 20
>>> encapsulation frame-relay
>>> clock rate 2000000
>>> frame-relay map ip 150.100.100.2 502 broadcast
>>> frame-relay map ip 150.100.100.6 502
>>> no frame-relay inverse-arp
>>> frame-relay lmi-type cisco
>>> 
>>> R5 after reload:
>>> 
>>> interface Serial0/1/0
>>> bandwidth 64
>>> ip address 150.100.100.5 255.255.255.0
>>> ip bandwidth-percent eigrp 1001 43
>>> ip hello-interval eigrp 1001 20
>>> ip hold-time eigrp 1001 15
>>> encapsulation frame-relay
>>> clock rate 2000000
>>> frame-relay map ip 150.100.100.6 502
>>> frame-relay map ip 150.100.100.2 502 broadcast
>>> no frame-relay inverse-arp
>>> frame-relay lmi-type cisco
>>> end
>>> 
>>> A new line appears in the running config for the interface: “ip
hold-time
>>> eigrp 1001 15”
>>> When R5 comes back up, R5 continues to use a hold timer of 180 seconds for
R2,
>>> but R2 begins using a hold timer of 15 seconds for R5. This mismatch
results
>>> in a continuously flapping neighbor adjacency. 
>>> Attempting to add the command “ip hold-time eigrp 1001 180” to the
>>> serial0/1/0.256 interface on R2 does not have any effect and never shows
in
>>> the running configuration of the interface even after a reload.
>>> 
>>> I am able to clear the issue by removing the “hello-interval” and
>>> “hold-time” commands from the R5 (or R6) serial interface and then
>>> reapplying the “ip hello-interval eigrp 1001 20” command. However,
this
>>> condition only lasts until the next reload.
>>> 
>>> IOS versions for R2, R5 and R6 in this pod:
>>> R2: c2800nm-adventerprisek9-mz.124-24.T2
>>> R5: c2800nm-adventerprisek9-mz.124-15.T9
>>> R6: c2801-adventerprisek9-mz.124-24.T2
>>> 
>>> I am just not sure is this is an isolated issue, or if there is something
I am
>>> missing
>>> 
>>> Many thanks
>>> Kind regards
>>> jpn
>>> 
>>> _______________________________________________
>>> For more information regarding industry leading CCIE Lab training, please
visit www.ipexpert.com
>>> 
>>> Are you a CCNP or CCIE and looking for a job? Check out
www.PlatinumPlacement.com
>>> 
>>> http://onlinestudylist.com/mailman/listinfo/ccie_rs
>> _______________________________________________
>> For more information regarding industry leading CCIE Lab training, please
visit www.ipexpert.com
>> 
>> Are you a CCNP or CCIE and looking for a job? Check out
www.PlatinumPlacement.com
>> 
>> http://onlinestudylist.com/mailman/listinfo/ccie_rs
> 



_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com

Are you a CCNP or CCIE and looking for a job? Check out 
www.PlatinumPlacement.com

http://onlinestudylist.com/mailman/listinfo/ccie_rs

Reply via email to