Yes you should advertise the Node-SID with the algorithm you want to use for this prefix On top you should advertise the algorithm in the capability TLV for ISIS/OSPF
From: Veerendranatha Reddy Vallem <[email protected]<mailto:[email protected]>> Date: Sunday 28 September 2014 09:02 To: Wim Henderickx <[email protected]<mailto:[email protected]>>, "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: RE: [spring] Regarding SID/Lable distributed by IGPs Hi Henderick, Thanks for your reply. If I use OSPF/ ISIS SID in my label/SID stack for data forwarding, by default the label represents shortest path for the given prefix/node? If I use different algorithm for the same label/SID, I must notify which algorithm I used to for this label/SID while constructing label stack for specific path? Regards, Veerendranath From: Henderickx, Wim (Wim) [mailto:[email protected]] Sent: 28 September 2014 14:53 To: Veerendranatha Reddy Vallem; [email protected]<mailto:[email protected]> Subject: Re: [spring] Regarding SID/Lable distributed by IGPs At this stage we defined the solution such that a given SID/Label is associated with 1 algorithm. From: Veerendranatha Reddy Vallem <[email protected]<mailto:[email protected]>> Date: Sunday 28 September 2014 08:47 To: Wim Henderickx <[email protected]<mailto:[email protected]>>, "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: Re: [spring] Regarding SID/Lable distributed by IGPs Hi Henderick, Thanks for your reply. Yes, I found algorithm filed in TLVs. While constructing LSP , whether we need to provide the algorithm mapping information along with label/SID ? I think for construction of label stack ,it required to know what label represents (whether it is shortest path or best path to the given node/prefix), so we should have algorithm mapping information also along with label in SIDB. Please let me know if my understanding is wrong . Whether we can use multiple algorithms to represent same Node SID ? Regards, Veerendranath From: Henderickx, Wim (Wim) [mailto:[email protected]] Sent: 28 September 2014 14:29 To: Veerendranatha Reddy Vallem; [email protected]<mailto:[email protected]> Subject: Re: [spring] Regarding SID/Lable distributed by IGPs In the OSPF/ISIS drafts on SR we scoped an algorithm field which could be used for this purpose. So besides shortest path you could use a different algorithm. From: Veerendranatha Reddy Vallem <[email protected]<mailto:[email protected]>> Date: Sunday 28 September 2014 08:15 To: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: [spring] Regarding SID/Lable distributed by IGPs Hi, I have a small query related to SID distributed by IGPs in the SR. Can anybody help me to clarify this. The SID/Label distributed by IGPs (OSPF or ISIS) for the prefixes/Nodes should represent Shortest Path to the given prefix or node? We can use same Labels for other representation like Best Effort Path also ? (The label distribution done by OSPF, but the next-hop for the node is calculated by other IGPs based on administrative decision) Regards, Veerendranath
_______________________________________________ spring mailing list [email protected] https://www.ietf.org/mailman/listinfo/spring
