04-17-2022 07:40 AM
Hello
I want to configure ISIS protocol with metric based on delay or bandwidth+delay.
I read a document which is not cisco official, below is the point that I saw:
IS-IS uses an arbitrary cost for its metric. IS-IS additionally has three
optional metrics: delay, expense, and error. Cisco does not support
these optional metrics
Does someone can point me to the right cisco reference documentation which can confirm or not Cisco does not support delay, expense and error as metric for ISIS ?
Thank you in advance for your reply
04-17-2022 04:52 PM
Hi
On this Book "implementing cisco ip routing (route) foundation learning guide" Cisco mention those metrics. But, this is old stuff and maybe they are not supporting anymore, but they did.
04-17-2022 11:20 PM
Hello,
here is another link to a Cisco Live presentation (slide/page 79) where this is mentioned...
https://www.ciscolive.com/c/dam/r/ciscolive/us/docs/2019/pdf/BRKRST-2315.pdf
05-03-2022 01:13 AM
Hello all
Thank you to @Georg Pauwen and @Flavio Miranda for their reply and sharing.
I want to provide more information about what I'm trying to acheave.
My goal is ISIS use also delay to make routing decision.
Find attach all files configuration and my lab network.
Into my lab, I have 2 Cisco ASR 920 (IOS XE 17.6.3) and 1 NCS 540 (IOS XR 7.4.1), MPLS, LDP, ISIS are running fine.
I deliver E-LINE service between lab01-2-cpe and lab01-3-cpe working fine.
I used feature performance-measurement (TWAMP) in order to push delay parameter over the link (lab01-2-asr920 <---->lab01-3-asr920, lab01-2-asr920 <----> lab01-1-ncs540 and lab01-1-ncs540 <------> lab01-3-asr920)
See below ISIS database when performance-measurement (TWAMP) is not active and when it is active.
ISIS database when TWAMP is not active (there is no delay information) | ISIS database when TWAMP is active (there is delay information) |
RP/0/RP0/CPU0:lab01-1-ncs540#sh isis database verbose IS-IS CORE (Level-2) Link State Database Total Level-2 LSP count: 3 Local Level-2 LSP count: 1 | RP/0/RP0/CPU0:lab01-1-ncs540#sh isis database verbose IS-IS CORE (Level-2) Link State Database Total Level-2 LSP count: 3 Local Level-2 LSP count: 1 |
Question: Does in this case delay information are compute with default metric to best route ?
If you see deeply the ISIS database when TWAMP is active, delay to go from lab01-2-asr920 (loopback 10.10.10.2) to lab01-3-asr920 (loopback 10.10.10.3) is higher than going through lab01-1-ncs540, I expected in terms of routes lab01-2-asr920 in order to go to lab01-3-asr920 choose --> lab01-1-ncs540 then lab01-3-asr920, but it is not working like that, ISIS is still using only default metric then in this case, I have to routes from lab01-2-asr920 to reach lab01-3-asr920 (loopback 0)
lab01-2-asr920#sh ip route 10.10.10.3
Routing entry for 10.10.10.3/32
Known via "isis", distance 115, metric 40, type level-2
Redistributing via isis CORE
Last update from 192.168.255.3 on TenGigabitEthernet0/0/26, 21:03:22 ago
Routing Descriptor Blocks:
* 192.168.255.6, from 10.10.10.3, 21:03:22 ago, via TenGigabitEthernet0/0/24
Route metric is 40, traffic share count is 1
192.168.255.3, from 10.10.10.3, 21:03:22 ago, via TenGigabitEthernet0/0/26
Route metric is 40, traffic share count is 1
lab01-2-asr920#sh ip cef 10.10.10.3
10.10.10.3/32
nexthop 192.168.255.3 TenGigabitEthernet0/0/26
nexthop 192.168.255.6 TenGigabitEthernet0/0/24 label 24000-(local:22)
lab01-2-asr920#
What I can do in order ISIS take into account delay information already present in his database for routing decision ?
Any suggestion will be wellcome
Thank you.
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