[j-nsp] EXP based LSP selection
Adam Vitkovsky
Adam.Vitkovsky at gamma.co.uk
Mon Aug 15 06:59:23 EDT 2016
Hi Dragan,
> Dragan Jovicic [mailto:draganj84 at gmail.com]
> Sent: Sunday, August 14, 2016 11:08 PM
>
>> Hi,
> >But since the policy is matching everything, then all routes in inet.3 will be
>> altered right?
>
> More importantly all the routes in mpls.0. Since this is most likely a bgp-free
> router it could very well not use inet.3. But you can be more specific in your
> policy, see bellow.
>
Yeah good point indeed.
>> I suspect that this is how the inet.3 table looks like for all the prefixes in it.
>> But then I'm wondering how come the return traffic for ping to 120.10.255.1
>> is not caught by the policy and looped back? Do you know what I mean
>> please?
>> Or is there some kind of split horizon rule, i.e. if packet cam through
>> interface-A it can't be sent out via int-A even if inet.3 says it should?
>
> I understand. No worries, there can be no loop. The policy is forced only for
> those LSPs which are best path for the specific prefix/label only. In our case
> the policy does not even consider two LSPs for returning traffic.
Excellent,
Now it clicked finally,
Just realized that whilst this wouldn't work in IP forwarding, in MPLS it just facilitates the LSP build up.
LSPs are unidirectional so the policy basically just helps to bind the local LDP label with the remote LDPoRSVP label (instead of just the LDP+IGP label).
> So I added two more LSPs for returning traffic in other direction:
>
> # run show route table inet.3
>
> inet.3: 20 destinations, 26 routes (4 active, 0 holddown, 20 hidden)
> + = Active Route, - = Last Active, * = Both
>
> 120.10.255.1/32 *[RSVP/7/1] 2d 08:29:37, metric 1000
> > to 120.10.50.2 via ge-1/0/9.500, label-switched-path lsp-be
> to 120.10.50.2 via ge-1/0/9.500, label-switched-path lsp-realtime
> [LDP/9] 11:41:34, metric 1000
> > to 120.10.50.2 via ge-1/0/9.500, label-switched-path lsp-be
> to 120.10.50.2 via ge-1/0/9.500, label-switched-path lsp-realtime
> 120.10.255.5/32 *[RSVP/7/1] 00:01:47, metric 1000
> to 120.10.110.2 via ge-1/0/9.1100, label-switched-path lsp-be-
> return-traffic
> > to 120.10.110.2 via ge-1/0/9.1100, label-switched-path lsp-
> realtime-return-traffic
> [LDP/9] 00:00:13, metric 1000
> to 120.10.110.2 via ge-1/0/9.1100, label-switched-path lsp-be-
> return-traffic
> > to 120.10.110.2 via ge-1/0/9.1100, label-switched-path lsp-
> realtime-return-traffic The returning traffic uses another set of LSPs trough
> our transit core.
> Modified policies:
>
>
> # show policy-options policy-statement lsp-policy1 from rib mpls.0; then cos-
> next-hop-map cos-map1; # show class-of-service forwarding-policy next-
> hop-map cos-map1 {
> forwarding-class BE {
> lsp-next-hop [ lsp-be lsp-be-return-traffic ];
> }
> forwarding-class REALTIME {
> lsp-next-hop [ lsp-realtime lsp-realtime-return-traffic ];
> }
> }
> Test:
>
> # run ping 120.10.255.1 source 120.10.255.5 count 200 rapid tos 160 PING
> 120.10.255.1 (120.10.255.1): 56 data bytes
> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> !!!!!!!!!!!!
> --- 120.10.255.1 ping statistics ---
> 200 packets transmitted, 200 packets received, 0% packet loss round-trip
> min/avg/max/stddev = 0.613/0.738/4.228/0.305 ms
>
> # run show mpls lsp statistics
> Ingress LSP: 4 sessions
> To From State Packets Bytes LSPname
> 120.10.255.1 150.1.3.1 Up 0 0 lsp-be
> 120.10.255.1 150.1.3.1 Up 200 17600 lsp-realtime
> 120.10.255.5 150.1.3.1 Up 0 0 lsp-be-return-traffic
> 120.10.255.5 150.1.3.1 Up 200 17600 lsp-realtime-return-traffic
>
>
> # run ping 120.10.255.1 source 120.10.255.5 count 200 rapid tos 0 PING
> 120.10.255.1 (120.10.255.1): 56 data bytes
> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> !!!!!!!!!!!!
> --- 120.10.255.1 ping statistics ---
> 200 packets transmitted, 200 packets received, 0% packet loss round-trip
> min/avg/max/stddev = 0.588/0.742/2.125/0.225 ms
>
> # run show mpls lsp statistics
> Ingress LSP: 4 sessions
> To From State Packets Bytes LSPname
> 120.10.255.1 150.1.3.1 Up 200 17600 lsp-be
> 120.10.255.1 150.1.3.1 Up 200 17600 lsp-realtime
> 120.10.255.5 150.1.3.1 Up 200 17600 lsp-be-return-traffic
> 120.10.255.5 150.1.3.1 Up 200 17600 lsp-realtime-return-traffic
>
Thank you so much buddy.
This is exactly the behaviour I need.
adam
Adam Vitkovsky
IP Engineer
T: 0333 006 5936
E: Adam.Vitkovsky at gamma.co.uk
W: www.gamma.co.uk
This is an email from Gamma Telecom Ltd, trading as “Gamma”. The contents of this email are confidential to the ordinary user of the email address to which it was addressed. This email is not intended to create any legal relationship. No one else may place any reliance upon it, or copy or forward all or any of it in any form (unless otherwise notified). If you receive this email in error, please accept our apologies, we would be obliged if you would telephone our postmaster on +44 (0) 808 178 9652 or email postmaster at gamma.co.uk
Gamma Telecom Limited, a company incorporated in England and Wales, with limited liability, with registered number 04340834, and whose registered office is at 5 Fleet Place London EC4M 7RD and whose principal place of business is at Kings House, Kings Road West, Newbury, Berkshire, RG14 5BY.
---------------------------------------------------------------------------------------
This email has been scanned for email related threats and delivered safely by Mimecast.
For more information please visit http://www.mimecast.com
---------------------------------------------------------------------------------------
More information about the juniper-nsp
mailing list