I've used MySQL to do this recently, and was able to connect OpenBD to
the MySQL cluster.
Make sure you're running the latest version of the MySQL driver, and set
up a custom data source inside OpenBD ("other" from the drop-down list).
The following entries in the data source fields worked with my MySQL
cluster (replace brackets with your own info):
JDBC URL: jdbc:mysql:loadbalance://[server 1 ip]:[server 1 port],[server
2 ip]:[server 2 port],[server 3 ip]:server 3 port/[database
name]?loadBalanceBlacklistTimeout=5000
Driver class: com.mysql.jdbc.Driver
Username and password should allow DB access across all SQL nodes in
your cluster.
In testing, when one of the nodes would fail, the pages would
occasionally take a while to load due to the failover timeout. You can
control the time it takes a node to be considered as "failed" by
adjusting the timeout value above, however, don't set it too low or you
could end up calling it failed when it's just under high load.
I was able to get a 2-node mysql cluster working, but, really, it
sucked. I'd go with at least 3 nodes. The more the better!
Hope this helps.
Warm regards,
Jordan Michaels
Vivio Technologies
http://www.viviotech.net/
Open BlueDragon Steering Committee
Railo Community Distributions
Skellington wrote:
Hello,
So I would like to add a secondary datasource to my application, for
failover purposes. Are there any good suggestions or examples on
setting up a secondary/failover datasource if the primary can not be
reached?
Thanks,
Charlie
--
Open BlueDragon Public Mailing List
http://www.openbluedragon.org/ http://twitter.com/OpenBlueDragon
mailing list - http://groups.google.com/group/openbd?hl=en
!! save a network - please trim replies before posting !!
