Neeraj -
One idea might be to use both a client ID field and a record ID field to
uniquely identify records. This approach has benefits (easy to add
clients, doesn't require as much client setup or planning) and a
potentially big drawback (no one unique record identifier).
Another would be to borrow a technique used in multi-master replication,
even if you're not using replication, to have your client databases
assign unique IDs. Benefit is unique identifier, drawback is you need
to plan ahead a bit more and do some setup on each client. See the
article at:
http://dev.mysql.com/tech-resources/articles/advanced-mysql-replication.html
and pay particular attention to the use of auto_increment_increment and
auto_increment_offset variables.
Hope this helps,
Dan
Neeraj wrote:
Hi all..
I have a scenario of many client databases and one central database..
Is there any way to get a unique ids across all databases (by any way) ,
because records from all the databases are finally get stored on a central
database. And the ids of transaction (a record) must remain unique in
central database.
These transaction ids are getting generated on client database..
Thanks in advance..
Cheers.....:-)
Neeraj Black Bits
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
