If you migrate to Oracle 11, with scan DNS entries defined (google "Oracle SCAN", and click the first link), then the connection string is simpler - just one ADDRESS entry.
In any case, you should separate configuration from Java code, define RAC configure in Oracle client configuration "tnsnames.ora" and then in resin configuration, define <database> within it, <jndi-name>, <driver>, etc... So your java code should just reference the datasource by JNDI name. From: [email protected] [mailto:[email protected]] On Behalf Of Steve Francis Sent: Friday, July 22, 2011 10:54 AM To: [email protected] Subject: [Resin-interest] Database Pooling I'm new to using Resin's database pooling, so I'm wondering if there is anything special about setting it up to use an Oracle RAC . I know that in my current java programs, the connect string is rather complex... databaseconnect=(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(Host=adcracdbq1-racvip.hiw. com)(Port=1521))(ADDRESS=(PROTOCOL=TCP)(Host=adcracdbq2-racvip.hiw.com<http://adcracdbq2-racvip.hiw.com>)(Port=152 1))(LOAD_BALANCE=yes)(FAIL_OVER=NO)(CONNECT_DATA=(SERVER=DEDICATED)(SERVICE_NAME =HMIQ))) How is the RAC environment configured with Resin? Thanks, Steve Francis Technical Adviser -- Application Technologies IHG Office: 770-442-7157 Cell: 770-906-3122 IM: francisihg
_______________________________________________ resin-interest mailing list [email protected] http://maillist.caucho.com/mailman/listinfo/resin-interest
