[c-nsp] manually crafted bypass LSPs

[James Bensley]

On 16 October 2017 at 09:42,  <adamv0025 at netconsultings.com> wrote:
> Hi folks,
>
> Anyone tried running crafted bypass LSPs?
> I can't get mine working in the lab on 15.1F6 and have no idea why,
>
> Example:
> To primary LSP tail-end IP:
> set protocols rsvp interface ge-0/0/0.0 link-protection bypass test-to-pe1
> to 10.0.0.21
> set protocols rsvp interface ge-0/0/0.0 link-protection bypass test-to-pe1
> bandwidth 1m
> set protocols rsvp interface ge-0/0/0.0 link-protection bypass test-to-pe1
> path 10.0.0.2 loose
> set protocols rsvp interface ge-0/0/0.0 link-protection bypass test-to-pe1
> path 10.0.0.14 loose
> set protocols rsvp interface ge-0/0/0.0 link-protection bypass test-to-pe1
> path 10.0.0.20 loose
> To node protecting MP IP:
> set protocols rsvp interface ge-0/0/0.0 link-protection bypass test-to-cr2
> to 10.0.0.13
> set protocols rsvp interface ge-0/0/0.0 link-protection bypass test-to-cr2
> bandwidth 0
> set protocols rsvp interface ge-0/0/0.0 link-protection bypass test-to-cr2
> path 10.0.0.2 loose
> set protocols rsvp interface ge-0/0/0.0 link-protection bypass test-to-cr2
> path 10.0.0.8 loose
>
> Even with BW=0,
> The LSPs or rsvp interface extensive output keeps on complaining there's no
> sufficient BW (but for the automatic bypass LSP) -however I want my primary
> LSPs to use these crafted bypasses instead of the automatically created
> ones.
> (the bypass test-to-cr2 is practically the same thing as what would have
> bene created automatically)
> show rsvp session name BE_POP1_TO_PE1POP4_RED_0 extensive
> Type: Protection down
>       1 Oct 16 09:10:05 No suitable bypass with sufficient bandwidth[23
> times]
> show rsvp interface ge-0/0/0.0 extensive
> Protection: On, Bypass: 0, LSP: 1, Protected LSP: 0, Unprotected LSP: 1
>      10 Oct 15 19:41:24 No suitable bypass with sufficient bandwidth
> Bypass->10.1.0.40->10.1.0.17[4 times]
>
> LSP is configured with node/link protection:
> set protocols mpls label-switched-path BE_POP1_TO_PE1POP4_RED_0
> node-link-protection
>
> adam

Wrong list dude :)

Cheers,
James.