Re: [OSPF] I-D Action: draft-ietf-ospf-link-overload-05.txt

2017-04-20 Thread Ketan Talaulikar Talaulikar (ketant)
Hi Shraddha, The well-defined mechanism of getting interface-ids in RFC 4203 that you refer to is specific to TE applications since it is defined for the GMPLS/TE use-case in that RFC. It is not a generic mechanism since it requires use of TE LSAs and hence not clean option for use-cases where

Re: [OSPF] I-D Action: draft-ietf-ospf-link-overload-05.txt

2017-04-20 Thread Acee Lindem (acee)
Hi Shraddha, On 4/20/17, 12:46 AM, "Shraddha Hegde" wrote: >Hi Acee, > >The draft does not mandate use of RFC 4203. There are no MUST statements >associated with the recommendation. I don’t see any need to reference RFC 4203 since the Sub-TLV is sufficiently defined

[OSPF] draft-ietf-ospf-link-overload-06 - DR migration

2017-04-20 Thread Alexander Okonnikov
Hi authors, In case when the node that has the link to be overloaded is DR (for broadcast/NBMA link case), taking this link out of service could be disruptive. What if to modify procedure in such manner that when BDR receives Link-Overload-sub-TLV from DR, it generates Network LSA in advance,

Re: [OSPF] I-D Action: draft-ietf-ospf-link-overload-05.txt

2017-04-20 Thread Shraddha Hegde
Ketan, OSPF link overload has relevance for TE based applications as well as non-TE applications. So what's the problem in referring a well-defined mechanism of getting the interface-ids? Can you explain why you are so concerned with referencing a standard RFC which is out there and

Re: [OSPF] I-D Action: draft-ietf-ospf-link-overload-05.txt

2017-04-20 Thread Peter Psenak
Hi Shraddha, please see inline: On 20/04/17 08:46 , Shraddha Hegde wrote: Ketan, Pls see inline.. -Original Message- From: Ketan Talaulikar Talaulikar (ketant) [mailto:ket...@cisco.com] Sent: Thursday, April 20, 2017 10:06 AM To: Acee Lindem (acee) ; Acee Lindem

Re: [OSPF] I-D Action: draft-ietf-ospf-link-overload-05.txt

2017-04-20 Thread Ketan Talaulikar Talaulikar (ketant)
Shraddha, There are also other applications that your draft lists which are TE independent and hence the case for not referring to a specific way for signalling interface-ids which is TE specific. I don't understand why you would be so reluctant to remove a reference which is not even central

Re: [OSPF] I-D Action: draft-ietf-ospf-link-overload-05.txt

2017-04-20 Thread Shraddha Hegde
Ketan, Pls see inline.. -Original Message- From: Ketan Talaulikar Talaulikar (ketant) [mailto:ket...@cisco.com] Sent: Thursday, April 20, 2017 10:06 AM To: Acee Lindem (acee) ; Acee Lindem ; Shraddha Hegde Cc: ospf@ietf.org

Re: [OSPF] I-D Action: draft-ietf-ospf-link-overload-05.txt

2017-04-20 Thread Shraddha Hegde
Ketan, We do have traffic engineering applications that require link-overload functionality. Pls refer section 7.2. Rgds Shraddha -Original Message- From: Ketan Talaulikar Talaulikar (ketant) [mailto:ket...@cisco.com] Sent: Thursday, April 20, 2017 10:46 AM To: Shraddha Hegde