>I looked at that page in Doyle's book and I thought it's just a simple >mistake, or maybe IOS changed since he wrote that, but after reading this: > >>"Handling of default routes varies from protocol to protocol. RIP, IGRP, >>EIGRP and BGP automatically redistribute default routes while OSPF and IS-IS >>require you to explicitly advertise them with the default-information >>originate statement in your router configuration." > >I start to think that these folks (Doyle included) have some basic problem >with simple terminology.
Actually this terminology might not be that simple after all, as it is not clear whether something was redistributed unless it is advertised to peers. So revised statement below. Also, I just realized that my above sentence sounds less respectful than I intended, so I thought I would mention that I learnt a lot from "Routing TCP/IP" and it's probably the most useful networking book I've ever read. > There is no way "BGP automatically redistributes default routes". > However it's true that OSPF and ISIS don't advertise the default route > even if it's redistributed into them, unless 'default-information > originate' is specified. > >So the statement should be something like "RIP, IGRP, EIGRP and BGP >automatically advertise default routes that are redistributed into them, >while OSPF and ISIS require the 'default-information originate' >statement." At least I hope that this is a true statement... :) Probably this one is better: "RIP, IGRP, EIGRP and BGP automatically accept default routes that are redistributed into them, while OSPF and ISIS silently reject the redistributed default route unless 'default-information originate' is configured." Thanks, Zsombor >Btw, where is the Cisco Press errata page? > >Thanks, > >Zsombor > > > > > >At 06:57 PM 7/16/2003 +0000, Priscilla Oppenheimer wrote: >>wj chou wrote: >> > >> > do you guys know of any newer and good book talking about >> > redistribution and default routes? thanks! >> >> >>There's nothing wrong with the age of Doyle's Routing TCP/IP Volume I. The >>poster who claimed that didn't know the difference between a second edition >>and a second volume. Doyle published a second volume because he had more to >>say (on BGP, IPv6, multicast). He didn't publish it because his first volume >>was out-dated or wrong. >> >>A second edition updates a book. A second volume supplements it. Doyle >>hasn't published a second edition because there's no need for one yet. How >>much have routing protocols changed in the last few years? Not much. Plus if >>you can believe the errata at Cisco Press's site for Doyle Volume 1, there >>were only 2 mistakes in the book!? >> >>(I think that's a mistake, as in they forgot to publish the rest of the >>pages. Although Doyle's book is terrific, I doubt that it really only has 2 >>mistakes? But maybe. Check with Cisco Press. See if you can get them to >>publish a better errata....) >> >>Anyway, I believe what you are questioning is this statement from Doyle, >>volume 1: >> >>"After a default route is identified in the routing table, RIP, EIGRP, IGRP >>will automatically advertise it." >> >>There's a good paper on redistribution at CertificationZone, written by a >>frequent contributor to this list, John Neiberger, and Ron Trunk. >> >>The paper agrees with Doyle: >> >>"Handling of default routes varies from protocol to protocol. RIP, IGRP, >>EIGRP and BGP automatically redistribute default routes while OSPF and IS-IS >>require you to explicitly advertise them with the default-information >>originate statement in your router configuration." >> >>I suggest you send us your config. There's probably some weird wrinkle >>related to network classes or something. You may want to start a new thread. >>People that use the Web site don't look at messages that are this old >>sometimes.... >> >>Priscilla Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=72430&t=72211 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
