Oracle Distributed Systems
by Charles Dye


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.

If you have technical questions or error reports, you can send them to 
booktech@oreilly.com. (Please specify the printing date of your copy.)


This page was updated May 7, 2001.


Here's a key to the markup:

        [page-number]: serious technical mistake
        {page-number}: minor technical mistake
        <page-number>: important language/formatting problem
        (page-number): language change or minor formatting problem
        ?page-number?: reader question or request for clarification


UNCONFIRMED errors and comments from readers:


{44-46} This is not so much an error as a failure to understand on my part.
Fig 2-2, for which the preceding text is described as "the relevant multi_
threaded server INIT.ORA parameters," probably should include an eggman box
as well as walrus. It is not clear how the two listener addresses give rise
to the three listeners and one host of the diagram.

In figure 2-3, should that be two distinct instances pointing at the the
same database per the definitions of instance and and database given on
page 4?


{50} There is a discussion of cman.ora. Where (what directory) does this
usually reside in? Is this file required? The text says it is needed to
"customize."


{52} Paragraph 3:  How does [The ADDRESS section] tell the connection
manager to connect to the listener that is listening on port 1521? The
later example of different protocols makes it seem unlikely to be the
second ADDRESS section.


{62} In the last paragraph, the word "TRANACTIONS" is probably a
misspelling. Should be "TRANSACTIONS".


{66} "SnapShot_refresh_process" section probably should say "see
Job_queue_process".


?81? Second point of "Commit Point Site" paragraph;
What is meant by "If two or more nodes have the same COMMIT_POINT_STRENGTH, 
the determination of the the commit point is not specified"? 

Particularly, what does "is not specified" mean?  Does it mean the transaction 
aborts, that you get an error message, or what?


[97]  Under heading "Disadvantages of the local view approach";
"Truly there are no disadvantages to this approach..."

While I agree that the local view approach offers the most advantages with 
the least disadvantages, one thing the user might want to consider is the 
possibility of performance loss on large indexed tables since the view will 
not have an index.


{151} DRDA and other acronyms are used without a preceeding
definition.


[173] The last line on the page now reads:

	FROM customers@PSLS.BIGWHEEL.COM;

Should read:

	FROM customers@PHQS.BIGWHEEL.COM;

There would be no need for a link to PSLS when the view is created at that 
location.


?174? The sentence after the query at the top of the page reads:

	"This query accesses the data on PHQS.BIGWHEEL.COM only indirectly."

I'm confused about the PHQS part, as the view creation and subsequent query 
refer only to PSLS.


{194-5} Table 9-3: In the objects created table, should "I_SNAP$_table_name"
be included? Is there a job created, as stated in the previous paragraph.
Shouldn't the job be shown in fig 9-1?


{208} Figure 9-2: No master site is included, and it seems to be more
significant here than in Figure 9-1.

I would like to suggest consistency in labeling objects in the diagrams that
are contrasting the different types of snapshots. "EMP_SNAP" (Figure 9-1) is
almost certainly the same as "Emp Snapshot User View" (Figure 9-2), but as a
reader, I have to wonder if they have been marked differently for any reason.
The use of consistent labels and selection of objects to include would focus
attention on the significant differences.


{209} Although the read only and updatable snapshot discussions included set
up instructions, that for multi-master has neither the instructions nor a
pointer to where the instructions are to be found in the book.


{232} Middle of second paragraph reads:

	"Snapshot refreshes perform updates by deleting the old record and
	reinserting it with its new values."

After highlighting this in my book, I reviewed the actual update code
included on page 204. There, rather than delete/insert, actual updates are
performed. Which is correct?


{254}  Step 4  second line in example code;
When adding a master site for replication, the book lists the master 
definition site PHQS.BIGWHEEL.COM instead of PSLS.BIGWHEEL.COM.


[254] In the procedure call in Step 4:

1. A space is missing after the EXECUTE command.

2. Now reads:
	
	master => PHQS.BIGWHEEL.COM

Should read:

	master => PHLS.BIGWHEEL.COM.

PHQS is the master definition site, where as we are adding a master site,
which in the case is PHLS.


{309} In item number 3, you must specify the "gname" item when your group
name and schema name are different, which they are in this example.


[310-312] The text on pages 310-312 has no reference to the ptex of Figure
13-1. Is this the same as the PSFO of Figure 13-2? Would a refresh of from
PHQS to PCAL occur at the same time as PTEX (PSFO?)


(315) Line 6 in the box:  "do" should be "to".


{347}  In the first example, in the call to dbms_repcat.make_column_group(), 
the first parameter is listed as "gname" (typically identifying
a replication group name); in fact it is a "sname" (a schema name).


{347}  Below second paragraph in example code;
The page shows an example of how to execute the 
dbms_repcat.make_column_group package but the first parameter listed shows 
GNAME and it should read SNAME.


{350, 351} The third line of code from the bottom of page 350 and the third 
and eighth lines of code from the top of page 351 now read:

	EXECUTE dbms_repcat.add_priority_site

Should read:

	EXECUTE dbms_repcat.add_site_priority_site