Junos
Junos

Segment Routing for Legacy LFA , About backup path calculation problem

‎06-06-2019 03:56 AM

Dear Expert,

 

I have a problem with Segment Routing for Legacy LFA. Backup Path calculates the problem. The topology diagram is as follows:

top01.png

 

 

Currently, the routing table inet.3 from R1 to R9.

[edit]
jcluser@R1# show protocols mpls |display set 
Jun 06 10:46:52
set protocols mpls icmp-tunneling
set protocols mpls interface ge-0/0/0.0
set protocols mpls interface ge-0/0/1.0
set protocols mpls interface ge-0/0/2.0
set protocols mpls label-range static-label-range 10000 20000

[edit]
jcluser@R1# show protocols isis | display set 
Jun 06 10:46:55
set protocols isis apply-groups gr-isis
set protocols isis reference-bandwidth 1000g
set protocols isis backup-spf-options remote-backup-calculation
set protocols isis backup-spf-options node-link-degradation
set protocols isis backup-spf-options use-source-packet-routing
set protocols isis source-packet-routing srgb start-label 1000
set protocols isis source-packet-routing srgb index-range 9000
set protocols isis source-packet-routing node-segment ipv4-index 401
set protocols isis source-packet-routing node-segment ipv6-index 601
set protocols isis level 2 wide-metrics-only
set protocols isis interface ge-0/0/1.0 node-link-protection
set protocols isis interface ge-0/0/1.0 level 2 lan-neighbor 0000.0000.0002 ipv4-adjacency-segment protected index 100
set protocols isis interface ge-0/0/2.0 point-to-point
set protocols isis interface ge-0/0/2.0 node-link-protection
set protocols isis interface ge-0/0/2.0 level 2 metric 2000
set protocols isis interface ge-0/0/2.0 level 2 ipv4-adjacency-segment unprotected dynamic
set protocols isis interface ge-0/0/2.0 level 1 disable
set protocols isis interface all node-link-protection
set protocols isis interface all level 1 disable

 

jcluser@R1# run show route table inet.3 192.168.0.9/32 
Jun 06 10:47:51

inet.3: 9 destinations, 9 routes (9 active, 0 holddown, 0 hidden)
+ = Active Route, - = Last Active, * = Both

192.168.0.9/32     *[L-ISIS/14] 00:16:54, metric 4000
                    > to 10.0.0.3 via ge-0/0/0.0, Push 1409
                      to 10.0.0.23 via ge-0/0/2.0, Push 1409

jcluser@R1# run show route table inet.3 detail 192.168.0.9 | match "entry|L-ISIS|weight|operation" 
Jun 06 10:48:42
192.168.0.9/32 (1 entry, 1 announced)
        *L-ISIS Preference: 14
                Next hop: 10.0.0.3 via ge-0/0/0.0 weight 0x1, selected
                Label operation: Push 1409
                Next hop: 10.0.0.23 via ge-0/0/2.0 weight 0xf000
                Label operation: Push 1409
[edit]
jcluser@R1# run show route forwarding-table destination 192.168.0.9/32 
Jun 06 10:49:15
Routing table: default.inet
Internet:
Enabled protocols: Bridging, 
Destination        Type RtRef Next hop           Type Index    NhRef Netif
192.168.0.9/32     user     0                    ulst  1048582    16
                              10.0.0.3           ucst      610    16 ge-0/0/0.0
                              10.0.0.23          ucst      605    12 ge-0/0/2.0

The first question: R1 to R9 I have already based load balancing on the packet, and also have two next hop addresses in the forwarding table. The routing table inet.3 shows the selected R4 as the next hop. If the traffic reaches 192.168.0.9, will it pass the address 10.0.0.23?

 

The second question: R3 I have set the interface to an unprotected state. When R1 calculates the backup path, R3 should not be selected. But it seems that the route seems to have chosen R3.

 

[edit]
jcluser@R3# show protocols isis | display set 
set protocols isis export isis-expoert
set protocols isis reference-bandwidth 1000g
set protocols isis source-packet-routing srgb start-label 1000
set protocols isis source-packet-routing srgb index-range 9000
set protocols isis source-packet-routing node-segment ipv4-index 403
set protocols isis source-packet-routing node-segment ipv6-index 603
set protocols isis level 2 wide-metrics-only
set protocols isis interface ge-0/0/1.0 point-to-point
set protocols isis interface ge-0/0/1.0 level 2 ipv4-adjacency-segment unprotected dynamic
set protocols isis interface ge-0/0/2.0 point-to-point
set protocols isis interface ge-0/0/2.0 node-link-protection
set protocols isis interface ge-0/0/2.0 level 2 metric 2000
set protocols isis interface ge-0/0/2.0 level 2 ipv4-adjacency-segment unprotected dynamic
set protocols isis interface ge-0/0/2.0 level 2 ipv6-adjacency-segment unprotected dynamic
set protocols isis interface ge-0/0/2.0 level 1 disable
set protocols isis interface all level 1 disable
set protocols isis interface all level 2 ipv4-adjacency-segment unprotected dynamic
[edit]
jcluser@R3# run show isis adjacency detail 
R6
  Interface: ge-0/0/1.0, Level: 2, State: Up, Expires in 20 secs
  Priority: 0, Up/Down transitions: 1, Last transition: 00:22:33 ago
  Circuit type: 3, Speaks: IP, IPv6
  Topologies: Unicast
  Restart capable: Yes, Adjacency advertisement: Advertise
  IP addresses: 10.0.0.25
  IPv6 addresses: fe80::250:56ff:fea2:755a
  Level 2 IPv4 unprotected Adj-SID:      18, Flags: --VL--

R1
  Interface: ge-0/0/2.0, Level: 2, State: Up, Expires in 26 secs
  Priority: 0, Up/Down transitions: 1, Last transition: 00:22:57 ago
  Circuit type: 2, Speaks: IP, IPv6
  Topologies: Unicast
  Restart capable: Yes, Adjacency advertisement: Advertise
  IP addresses: 10.0.0.22
  IPv6 addresses: fe80::250:56ff:fea2:c0d7
  Level 2 IPv4 unprotected Adj-SID:      16, Flags: --VL--
  Level 2 IPv6 unprotected Adj-SID:      17, Flags: F-VL--

Is it that I understand the problem with the protection mechanism? Or have an official description. thank.

 

4 REPLIES 4
Junos

Re: Segment Routing for Legacy LFA , About backup path calculation problem

‎06-06-2019 09:22 AM

I may not be right. This is a bit complicated topology and technology. Better to have JTAC or POC for end to end 

 

To answer your question:

1. Traffic won't be going through 10.0.0.23. The reason is, if you check the "weight" of the next-hop, they are different 

Weight 1 is most preferred. Weight has to be the same to ECMP. 

10.0.0.23 is pre-installed in case of failure. But won't be used for forwarding unless the other link fails 

192.168.0.9/32 (1 entry, 1 announced)
        *L-ISIS Preference: 14
                Next hop: 10.0.0.3 via ge-0/0/0.0 weight 0x1, selected
                Label operation: Push 1409
                Next hop: 10.0.0.23 via ge-0/0/2.0 weight 0xf000
                Label operation: Push 1409

 

2. In most of Traffic engineering mechanisms, like RSVP link-protection, fast reroute, the protection route/path is calculated per node. R1 calculates backup path based on its own configuration, for interface 1, it will find interface 2 to protect, regardless of downstream device. For R3, it doesn't need protection for interfaces, so ideally it should not install more than 1 next-hops. If you do "show route 192.168.0.9" on R3, what do you see?

 

 

 


Mengzhe Hu
JNCIE x 3 (SP DC ENT)
Junos

Re: Segment Routing for Legacy LFA , About backup path calculation problem

‎06-08-2019 09:26 AM

Dear mhu,

 

                    > to 10.0.0.25 via ge-0/0/1.0, Push 1409

[edit protocols isis]
jcluser@R3# run show route 192.168.0.9/32                 

inet.0: 27 destinations, 27 routes (27 active, 0 holddown, 0 hidden)
+ = Active Route, - = Last Active, * = Both

192.168.0.9/32     *[IS-IS/18] 08:10:32, metric 3000
                    > to 10.0.0.25 via ge-0/0/1.0

inet.3: 8 destinations, 8 routes (8 active, 0 holddown, 0 hidden)
+ = Active Route, - = Last Active, * = Both

192.168.0.9/32     *[L-ISIS/14] 08:10:32, metric 3000
                    > to 10.0.0.25 via ge-0/0/1.0, Push 1409
Junos
Solution
Accepted by topic author likang
‎06-10-2019 07:53 AM

Re: Segment Routing for Legacy LFA , About backup path calculation problem

‎06-10-2019 07:44 AM

Hi Likang,

 

This looks correct to me. On R3, you are only seeing one valid next-hop. According to your configuration, you don't want protection. The route/next-hop matches your configs 

 

 


Mengzhe Hu
JNCIE x 3 (SP DC ENT)
Junos

Re: Segment Routing for Legacy LFA , About backup path calculation problem

‎06-10-2019 07:55 AM

Dear mhu,

 

Thank you for your support.

 

I have another question now, about the P and Q node space issues, I want to know how they are calculated. Thank you.