IP Routing By Ravi Malhotra The unconfirmed error reports are from readers. They have not yet been approved or disproved by the author or editor and represent solely the opinion of the reader. Here's a key to the markup: [page-number]: serious technical mistake {page-number}: minor technical mistake : important language/formatting problem (page-number): language change or minor formatting problem ?page-number?: reader question or request for clarification This page was updated December 30, 2002. UNCONFIRMED errors and comments from readers: [4] Near bottom of page; The following statement is not entirely true: "...if the next hop specified goes down, the router will remove the static route entry from the routing table" You can easily configure a static route to remain in the routing table, should the specified next hop go down by using the "permenent" keyword as follows: ip route 172.25.0.0 255.255.0.0 10.1.1.1 permenent (12) 3rd paragraph; 3.By exclusion, network 192.168.1.0, connected... ...NewYork will not process any RIP updates received on Ethernet0... Ethernet0 should be corrected as Ethetnet1, according to configuration sample on page 11, IP address of Ethernet1 is 192.168.1.1. So 192.168.1.0 will not be processed on Ethernet0, not 1. {14} last 2 lines on the page; C 171.16.250.0 is directly connected, Serial0 C 171.16.252.0 is directly connected, Serial1 needs to be : C 171.16.251.0 is directly connected, Serial1 C 171.16.252.0 is directly connected, Serial0 so in the first line the ip address & the serial is wrong, and in the second line the serial is wrong. And i also think that the order of the two lines needs to change then, because i saw that in every example the output list is ordered on serial. {18} both routing tables; the lines C 172.16.1.9 is directly connected, Ethernet0 should read C 172.16.1.0 is directly connected, Ethernet0 in both tables. [19] Section "Fast Switching", Paragraph 2; Where does 142.148.185.192 come from? It does not show up in the routing table shown for the "Parallel Paths" section. Is this a real address which was in the text before you changed to private addresses in the examples? [21] Figure 2-3 and code block below it.; Dear Sirs, I have a minor problem understandint the current "errata". Please see below : -----quote from current errata------------------------- [21] Figure 2-3. Several changes in this diagram. On routers A, B and C change all interfaces from Ethernet to Serial: On router A change E1 and E2 to S0 and S1, respectively. On router B change E1 and E2 to S0 and S1, respectively. On router C change E1 and E2 to S0 and S1, respectively. -----quote from current errata------------------------- In the diagam, (Figure 2-3) There appears to be no E1 and E2 for router A , and the same goes for router B and router C. SO, how am I supposed to change the interfaces in the diagram, if E1 and E2 are not there ? The only reference to E1 and E2 for all three routers are in the text(code block) that goes with the Figure 2-3, and NOT in the diagram itself. {22} routing table at top of page; a few typo's here... the line C 172.17.1.9 is directly connected, Ethernet0 should read C 172.17.1.0 is directly connected, Ethernet0 towards the bottom of the table, the line routing via 172.16.250.2, Ethernet1 should read routing via 172.17.250.2, Serial0 (23) para 1; Test says example net is routers A, B, and C are all connected by ethernets, where as the routing tables shown on page 21 (para 4) and page 22 (para 2) show a combination of Ethernet and Serial ... thus are really for "TraderMary's network". [29] First full paragraph; RIP will not automatically carry the default route as mentioned on page 29. In fact, the example will not work as advertised. In order to force RIP to carry the configured default route, one must either redistribute static routes into RIP or use the "default-information originate" command. Without one of these two commands RIP will simply not advertise the statically configured default route. [32] anyware in chapter 2; 2 questions that this chapter might address: Since serial links are point to point, why use /28 subnet masks instead of /30? (eg. portland example in figure 2-7) Why advertise default routes via rip instead of just configuring them into the "non-core" routers? (eg. portland router in example on page 28) {41} last sentence of paragraph at top of page; the line ending with "all serial interfaces to be 1,544 kbps and 20,000 ms, respectively (see Table 3-1)." should read "all synchronous serial interfaces to be 1,544 kbps and 20,000 usec, respectively (see Table 3-1)." as per the errata for pages 40 and 41. [59] Multiple default routes section; I believe configuring eth0 on core1 and core2 with the same Ip address on net 192.168.1.0/24 will not work. I think you should probably have given core2 another address (like 192.168.1.2). P.S. You did not show the config of the extra serial interface S1 on the router branch1, but I found that implication pretty obvious to follow. (as shown by the second default route on Serial0) [88] Second line; The administrative distance show in this example is incorrectly shown as 90. In actual fact, manual EIGRP summary routes are given an administrative distance value of 5, by default. Of course, an administrative distance of 90 could be assigned to a manual EIGRP summary route by adding the (optional) admin-distance to the end of the command: ip summary-address eigrp 10 172.26.0.0 255.255.240.0 90 However, this option was not available until 12.0(7)T... (and not discussed in the book) {94} Bottom of the page; One of the more important new feature of RIPv2 isn't mentioned in the chapter introduction, multicast support. RIPv2 uses multicast updates via the well known address (224.0.0.9) whereas RIPv1 uses resource intensive broadcasts (255.255.255.255). [139] last line of page; In order to summarize 10.3.0.0/16 the summary command should be: summary-address 10.3.0.0 255.255.0.0 not summary-address 10.3.0.0 255.255.252.0 Also, the other errata entry for this command is incorrect... [210] Second full paragraph; The redistribution pitfall that is offered is not possible. For example, the sentence "It may appear to routers in domain 1 that (network) X is reachable via R2!" is incorrect. EIGRP maintains an administrative distance of 90, however, it assigns an administrative distance of 170 to routes that are external (i.e. redistributed routes). This means that routers in domain 1 will always prefer the internal route to net X over the external route to X learned via OSPF. EIGRP provides this feature to avoid routing loops in the first place. Perhaps the EIGRP section should discuss this feature as well?? Please see: http://www.cisco.com/warp/public/105/admin_distance.html [211] Middle of page; Although the maximum number of paths is 6, the default number is 4. The book implies that the default number of paths is 6 which is incorrect. To increase the number of paths from 4 to 6 requires an explicit command: maximum-paths 6