From: Jonathan V Hays (jhays@jtan.com)
Date: Wed May 14 2003 - 10:45:16 GMT-3
All,
Here is my list of possible errors, omissions, and a few comments for
the book "Cisco OSPF Command and Configuration Handbook" by William R.
Parkhurst, ISBN 1-58705-071-4, First Printing, April 2002. This is an
excellent book for obtaining an in-depth knowledge of OSPF and is a very
clear and well-written text, but like most computer books these days, is
full of errors.
I have merged my list with the official Cisco Press list for your
convenience. I'm sure some of you may have additional errors I missed or
perhaps can clarify something that I thought was an error and is really
a misunderstanding on my part. I welcome any feedback for this post.
The current official errata for the book is on the Cisco Press web site
at:
http://www.ciscopress.com/catalog/product.asp?product_id={88F14DBB-F999-
41E0-AAA5-93439DC7109A}
I suggest doing all the exercises in this book in your lab. Even the
ones that look too simple to bother with may surprise you with
unexpected lessons is the Cisco IOS. Do the easy ones for speed
practice.
In a number of the labs, the configuration change did not take effect
after either interface shutdown or "clear ip ospf processes" as
indicated in the book but required a router reboot to take effect. I
have not noted these errors since they may be IOS version dependent.
I hope this is of use to others in their study!
Jonathan Hays
--- Page 6 - Figure 1-2 has the leftmost and rightmost routers labeled as PE (Provider edge) but they should be labeled CE (Customer edge).--- Pages 34, 41, 49, 58 - Configuration for Router B lists the command "lockrate 64000" which should be "clockrate 64000" on each of these pages.
--- [from Cisco Press] Page 84 - Regarding the Router B routing table: rtrB#show ip route, the second line in the second half of the configuration should read:
O 1.1.1.1 [110/65] via 10.1.1.1, 00:01:01, Serial0
--- [from Cisco Press] Page 193 - Regarding Figure 6-3, the configuration that reads: Loopback 0 51.51.51.51/24, should be changed to Loopback 1
--- Page 210 - The output of "show ip access-lists 1" under Step 4 incorrectly displays both access list 1 and access list 2. It should only display access list 1.
--- Page 277 - The configuration for Router C has numerous errors: interface loopback0 ip address 2.2.2 255.255.255 **SHOULD BE: ip address 3.3.3.3 255.255.255.255
interface Ethernet0 ip address 172.16.1.2 255.255.255.0 **SHOULD BE: delete this interface since it is not shown on the diagram or seen in any of the outputs. It was probably pasted in from a previous scenario
interface serial0 ip address 10.1.1.2 255.255.255.252 encapsulation frame-relay frame-relay map ip 10.1.1.1 120 broadcast **SHOULD BE: ip address 10.1.1.6 255.255.255.252 frame-relay map ip 10.1.1.5 120 broadcast
router ospf 1 network 2.2.2.2 0.0.0.0 area 2 network 10.1.1.0 0.0.0.3 area 0 neighbor 10.1.1.1 **SHOULD BE: network 3.3.3.3 0.0.0.0 area 3 network 10.1.1.4 0.0.0.3 area 0 neighbor 10.1.1.5
--- Page 279 - I think the author should warn the reader about lack of routes in the routing tables. Changing the hello interval allows the routers to become neighbors, but doing this will not change the fact that Router B and Router C are OSPF network type NBMA and Router A is OSPF network type POINT-TO-POINT. Although the author does not specifically address this issue, the routing tables will be empty because of the OSPF network type mismatch. The extensive errors on Router C's configuration (page 277) makes me wonder if he meant to have additional commands configured on Router B and Router C.
--- Page 282 - Under the "show ip route" output the entry for network 172.16.0.0/24 should be deleted since there was no configuration given to produce that route in the routing table.
--- Page 289 - At the top of the page the author states that "The router with the lowest OSPF priority on a network will be elected the DR for the network." This is incorrect. The router with the HIGHEST priority (largest numerical value) will be elected DR, as he says at the bottom of page 429 of the book.
--- Pages 310, 320 - The word "ONLY" is used incorrectly under "Purpose:" on both pages. The third sentence is:
"This command will redistribute classful routes into OSPF only if the subnets keyword is not used."
The phrase "only if the subnets keyword is not used" in the above statement logically implies that if the "subnets" keyword *is* used then classful routes will NOT be redistributed. This is not true, since using the "subnets" keyword redistributes *both* classful and classless routes into OSPF.
I think the word "only" needs to be moved in front of the word "classful" to correct the sentence:
"This command will redistribute ONLY classful routes into OSPF if the subnets keyword is not used."
--- Page 310, 311, 322, 323 - The configurations shown on pages 310-311 and 322-323 are missing the "bandwidth 64" statements under the serial interfaces. Without those statements, you would not be able to achieve the metrics given in the "show ip route" outputs and Figure 14-2 in the book. Chapter 14 configurations on pages 330-332 correctly include the "bandwidth 64" statements.
Also on pages 311 and 322 Router B should have "clockrate 64000" under its Serial 1 interface.
--- Page 325 - The middle of Page 325 shows a metric of 200 for 10.1.1.4 but I think the metric should be 20. Perhaps this output was copied from the next page, which is the output after "set metric 200" has been applied in the route-map.
--- Page 345 - Regarding the "summary-address ip-address mask not-advertise" command, the author states "Using the not-advertise keyword suppresses the advertisement of the summary route by the ASBR or ABR." I think the sentence should say only "ASBR" and not include "ABR." In fact on page 348 he states "The summary-address command can only be used on an ASBR."
--- Page 375 - The output of the "show ip ospf neighbor detail" command includes "Database-filter all out" but I did not see this in the output. I could only get this output using "show ip ospf interface s1."
--- Pages 378, 387 - The output given under "debug ip ospf adj" is incorrect. The desired output can be achieved using the "debug ip ospf hello" command.
--- Page 383 - The "ip ospf flood-reduction' command is not available on the IOS that I used, 12.1(15) although the author states that it is available in 12.1.
--- Page 394 - The author omits referring to the frame switch configuration on page 266. While not an error, it would be a helpful addition and would be consistent with other sections that provide complete configurations.
--- Page 395 - Although not a functional error, for consistency Router C should have "frame-relay lmi-type ansi" as the other routers do.
--- Page 401 - In the middle of the page, the word "leaning" should be "learning" in the sentence "This is not necessary for the operation of OSPF over Frame Relay but only prevents the routers from leaning DLCIs that are not used."
--- Pages 413, 414 - Again, the router configurations are missing "bandwidth 64" which would be required to achieve the desired metrics shown.
--- [from Cisco Press] Page 415 - Regarding the Router B configuration contained in the top table, please add the following to the configuration:
frame-relay-map ip 10.1.1.3 110 broadcast
Regarding the Router C configuration contained in the top table, please add the following to the configuration: frame-relay-map ip 10.1.1.2 120 broadcast
--- Page 432 - This example is confusing to the reader and should be clarified with a sentence or two.
In the example, Router A and Router B are two OSPF routers on a common ethernet segment. Router A has the default OSPF priority of 1 and we see Router B (the current DR) configured with the "ip ospf priority 2" command. After Router B's ethernet interface is shut down Router A becomes the DR.
But this occurrence of Router A being elected DR has NOTHING to do with priority, since in fact Router B has the higher priority. Router B would become DR if both routers were booted at the same time because it has the higher priority (highest numerical priority value). Router A has become DR simply because the existing DR on the subnet has been lost and Router A is the only candidate. Without knowing this fact, and given the incorrect statement regarding priority on page 289, the reader may incorrectly infer that Router A has become the DR because it has a lower priority number than Router B.
This archive was generated by hypermail 2.1.4 : Mon Jun 02 2003 - 15:13:42 GMT-3