This archive is retained to ensure existing URLs remain functional. It will not contain any emails sent to this mailing list after July 1, 2024. For all messages, including those sent before and after this date, please visit the new location of the archive at https://mailman.ripe.net/archives/list/db-wg@ripe.net/
The router object
- Previous message (by thread): The router object
- Next message (by thread): The router object
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Antonio_Blasco Bonito
bonito at nis.garr.it
Thu Jul 21 11:57:04 CEST 1994
Tony, > > > bonito at nis.garr.it (Antonio_Blasco Bonito) writes: > > * > * The example should list also interior peers with internal routing protocols > * and the explanation text should mention that too. > * > Well it does have an IBGP interior example. I can add something in the > text as well. I meant IGP should be there too. Usually you get information on internal routes via some IGP (OSPF, I-ISIS, RIP, IGRP, ...). It is wise to mention that in the example, I guess. > * > > * > > * > <Interface Address> must be a "dotted-quad" represented > * > host address. It should be noted that at least ONE ifaddr > * > must be configured for the inet-rtr object to be valid. > * > This facilitates the registering of route servers which > * > may only have one interface address and are purely routing > * > engines. > * > * Uhmmm, a route server which does not route packets but is only used by > * actual routers as a source of routing information IS NOT a router. Does it > * need to be registered? If yes I think it should be clearly distinguishable > * by actual routers. > * > Uhmmm..sorry not where I come from. A router is something that > exchanges routing information - I see no reason why a router needs to > forward packets. In fact a route server is specifically configured to > not to forward packets. You are confusing routing with forwarding. I see no > real reason to distinguish an RS. Whilst it is perhaps a special case > it still fits simply into the general schema of the object. By having > a single interface you give an indication of route serving (of course > you can forward in and out of this as well but..). > It is not the intent of this object to catch every special case > (otherwise we have to support to ebgp multihop - tunnels, etc). OK, this is terminology and "network philosophy". Leave as it is. > > * > * Text to be added: > * Note that in some cases a router configured as being in more > * than one AS can also peer with itself to exchange routes > * among its ASes > * > Really ? what is the exact purpose of this and how is it acheived ? Suppose you are running two BGP processes in a router you may want to run an IBGP between the two to echange routes eventually applying some policy filtering to the exchange. Exactly what is done if you have two different boxes... > > * > > * > notify: > * > The notify attribute contains an email address to which notifi- > * > cations of changes to this object should be send. > * > * The meaning and usage of this attribute is not clear: which kind of changes > * ? > * > Any update - please look at the relavent ripe document for more > details. > ripe-096.txt btw. OK, so please insert a reference to ripe-096 in the text. > > * > > * > maintainer: > * > The maintainer attribute contains a registered maintainer name. > * > > * The meaning and purpose of this attribute is not clear. > > See ripe-096 for more details. OK, so please insert a reference to ripe-096 in the text. > > --Tony. > ---------- ---------- Antonio_Blasco Bonito E-Mail: bonito at nis.garr.it GARR - Network Information Service c=it;a=garr;p=garr;o=nis;s=bonito c/o CNUCE - Istituto del CNR Tel: +39 (50) 593246 Via S. Maria, 36 Telex: 500371 CNUCE I 56126 PISA Italy Fax: +39 (50) 904052 ---------- ----------
- Previous message (by thread): The router object
- Next message (by thread): The router object
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
[ db-wg Archives ]