Hi, Amir!
On 11/10/06, Servers24 Network <servers24@gmail.com> wrote:
I don't really
> know how this DBI-Link works, but the way Juan said, maybe I can have
> Liferay write user information to a PgSQL DB and have DBI-Link propagates
> the data (username, password, first_name, last_name, city, state, country,
> zipcode and ... ) inside Liferay's DB to all appropriate DBs.
> Also I will disable ANY modification to the above mentioned fields in other
> applications so that a user need to change for e.g his name in Liferay, and
> then it will take effect in other applications as well.
> So in this method, I will use a PgSQL server with 2 databases ( one for
> DBI-Link and one for Liferay ) and a MySQL server with maybe 10 databases
> for other applications. In this case, CAS will read authentication
> information from Liferay DB in PgSQL.
> Any comments or ideas ?!
For how DBI-Link works, maybe pgfoundry docs (like
http://pgfoundry.org/docman/view.php/1000045/48/IMPLEMENTATION.txt)
can give a good big picture. Reading the mentioned document, it seems
like there could be better performance when reading through DBI-Link
from other datasources than writing to several other datasources from
PostgreSQL "DBI-linked" tables (just guessing, there can be some other
factors involved, and, as always, experiences and benchmarks are
welcome). If this is true, maybe it would be better to have the other
10 DBs in PostgreSQL reading from the centralized datasource, but of
course that might not be practical in your case, if it means a lot of
code rewriting. Anyway, I hope this helps and any feedback will be
followed with interest! ;)
Greetings!
