http://download.oracle.com/docs/cd/E11882_01/server.112/e10595/ds_admin001.htm
“If the remote database enforces global naming, then you must use the remote database global database name as the name of the link.”

http://download.oracle.com/docs/cd/E11882_01/server.112/e10595/ds_concepts002.htm
“When you set the initialization parameter GLOBAL_NAMES to TRUE, the database ensures that the name of the database link is the same as the global database name of the remote database. For example, if the global database name for hq is hq.acme.com, and GLOBAL_NAMES is TRUE, then the link name must be called hq.acme.com. Note that the database checks the domain part of the global database name as stored in the data dictionary, not the DB_DOMAIN setting in the initialization parameter file… If you set the initialization parameter GLOBAL_NAMES to FALSE, then you are not required to use global naming. You can then name the database link whatever you want. For example, you can name a database link to hq.acme.com as foo.”

I think that this is another example where a test case, like you created, is valuable.

I have referred to
http://download.oracle.com/docs/cd/B19306_01/server.102/b14231/ds_admin.htm#sthref4108

Regards,
Anand

This is what you quoted:
“The name that you give to a link on the local database depends on whether the remote database that you want to access enforces global naming. If the remote database enforces global naming, then you must use the remote database global database name as the name of the link.”

The above is different from what I quoted:
http://download.oracle.com/docs/cd/B12037_01/server.101/b10759/statements_5005.htm
”If the value of the GLOBAL_NAMES initialization parameter is TRUE, then the database link must have the same name as the database to which it connects. If the value of GLOBAL_NAMES is FALSE, and if you have changed the global name of the database, then you can specify the global name.”

Also found in the documentation for the latest release:
http://download.oracle.com/docs/cd/E11882_01/server.112/e10592/statements_5005.htm

I believe that the quotes that I supplied from the documentation agree with your test – it is apparently the value of GLOBAL_NAMES in the local database, not the remote database that determines whether you can change the DB link name.

In reference to “creates a database link to another database”, can you have a look at the below blog and comment.I am little confused as my results are not matching with the oracle documentation.

http://aprakash.wordpress.com/2010/01/30/db-link-and-global-name/

Regards,
Anand

Hopefully newer versions of Oracle are smart enough to only enter into a distributed transaction if an INSERT, UPDATE, or DELETE are issued against the target database.

From:
http://www.jlcomp.demon.co.uk/faq/dblink_commit.html
“Why does it seem that a SELECT over a db_link requires a commit after execution ?” “Because it does!” – note that the versions listed in the article are 7.0 – 9.2, so there is a chance that the behavior might have changed for more recent releases.

From:
http://download.oracle.com/docs/cd/B19306_01/server.102/b14231/ds_txns.htm
“Note that if a distributed transaction is set to read-only, then it does not use undo segments. If many users connect to the database and their transactions are not set to READ ONLY, then they allocate undo space even if they are only performing queries.”

Possibly related:
http://coskan.wordpress.com/2008/01/16/if-you-access-a-database-link-in-a-session-then-the-link-remains-open-until-you-close-the-session-ora-02020/

Maybe someone else reading this blog article will offer advice. Based on prior experience, it is still necessary in Oracle 10g R2 to issue a COMMIT or ROLLBACK after selecting data through a database link (the COMMIT or ROLLBACK does not need to happen immediately).

Why is it required?

Regards,
Anand