The errata list is a list of errors and their corrections that were found after the product was released.
The following errata were submitted by our customers and have not yet been approved or disproved by the author or editor. They solely represent the opinion of the customer.
Color Key: Serious technical mistake Minor technical mistake Language or formatting error Typo Question Note Update
| Version |
Location |
Description |
Submitted by |
Date submitted |
| PDF |
Page 82
Datatypes paragraph |
The SQL Statement is lacking the TABLE keyword.
|
Anonymous |
Aug 21, 2013 |
| Printed |
Page 90
1st paragraph |
You claim that "NULL=NULL" will always evaluate to FALSE.
This is an oversimplification.
It _eventually_ evaluates to FALSE, but _immediately_ it evalutates to NULL. The difference becomes clear when you look at
NOT (NULL=NULL) which evaluates to NOT (NULL), then to NULL and eventually to FALSE as well.
In general this paragraph does not explain the ternary logic very well.
As a bonus, why not show what happens to "De Morgan's law" in the presence of NULLs (I confess to get bitten by that from time to time).
|
Henryk Gerlach |
Apr 24, 2012 |
| Printed |
Page 93
Top |
Text on P. 92 describes a B*-tree as having blocks that contain index values, and the leaf nodes as having the row id for the associated row for ALL index values.
However, in the diagram on P.93, none of the rows in the branch-block index values are shown in the leaves. For example, 'Davis' is in a branch-block but in no leaves. Likewise for ALL the other index values in the branch-blocks - a serious, and confusing gaffe.
No-one seems to care about errata (all errata here are unconfirmed 3 years after release of the book), but hopefully someone will actually review these one day.
|
Bruce Wood |
Jun 12, 2010 |
| PDF |
Page 97
3rd from the bottom |
"the maximum number of partitions increased from 64 KB - 1 to 128 KB -1"
a number of partitions measured in a size measurement...? Maybe I am mis-reading this but that doesn't make any sense to me.(except if it is an index size limit for the partition or some such?)
|
Anonymous |
Apr 07, 2010 |
| Printed |
Page 161
6th par, half way down |
"and each file is a part of only one tablespace, as described in Chapter 4".
I think this should read "....Chapter 2"
Tablespaces and their relationships to datafiles are best described on page 34, Chapter 2 (as given in the index).
|
Andrew Ircha |
Dec 01, 2013 |
| Printed |
Page 190,191
|
You explain it IMHO correctly "serializable" in an academic DB
context (cf. http://en.wikipedia.org/w/index.php?title=Serializability&oldid=486257418):
p. 190 "As the names implies, serializable transactions appear as though the have been executed in a series of distinct,
ordered transactions."
Unfortionally oracle guarantees less with isolation_level=serializable: p.191 "This means that every statement within a transaction
will get the same consistent view of the date as it existed at the start of transaction".
While this might be usually enough, it is less than
"serializable" in the academic sense above (eg. http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:3233191441609).
You should call this "marketing bluff" and warn your readers more explicitely.
|
Henryk Gerlach |
Apr 24, 2012 |
| PDF |
Page 194
4. and 5. of 'A Simple Write Operation' |
Original:
4. The server writes the old image of the data to the redo buffers in memory, and then writes the changes to a rollback segment and modifies the employee data, which includes writing the SCN to the ORA_ROWSCN pseudocolumn in Oracle Database 10g or newer database releases.
5. The server process writes the redo buffers to disk, and then writes the rollback segments and the changed data to disk. The rollback segment changes are part of the redo, since the redo log stores all changes coming from the transaction.
Question:
Old image of the data should be written in the rollback buffer in memory. Changes should be written in redo log buffer in memory. Modifications are made in data buffer in memory. Am I right?
Did you swapped redo log buffer and rollback buffer by mistake, as in paragraph 5?
Can you please explain this more?
Thank you very much.
|
Nenad Dobrilović |
Feb 16, 2010 |
| Printed |
Page 220
Fig. 9-8 |
'analog' instead of 'along' (?)
|
Andrzej Kosowski |
Dec 06, 2008 |
| Printed |
Page 235
Multimedia |
'knows' instead of 'known'
|
Andrzej Kosowski |
Dec 06, 2008 |
| Printed |
Page 297
Node 1 in Fig. 12-4 |
|
Andrzej Kosowski |
Dec 06, 2008 |