I just discovered TRACE_LEVEL_FILE. I set that to 4 and see that in fact the connections are being rolled back :(
I'll have to investigate some more to see who is dropping the ball on the commit. On Monday, March 26, 2012 5:06:43 PM UTC-5, steve.ebersole wrote: > > I am having an issue using H2 (1.2.145) in in-memory mode with JTA. This > is part of the Hibernate testsuite. Initially the JTA tests against H2 > were erroneously passing because the set up was incorrectly using > auto-commit. So I tried I manually setting the auto-commit of Connections > obtained from the driver manager to false, at which point the tests started > failing. The manner in which they failed indicate that data committed in > one transaction was not visible in a second transaction as if the database > was dropped between. > > Note that this is NOT using H2 JTA/XA support. It is simply using H2 > connections obtained via DriverManager wrapped in an Enhydra DataSource. > The url is 'jdbc:h2:mem:db1;DB_CLOSE_DELAY=-1;MVCC=TRUE'. > > Anyone see anything obvious? If it weren't for the 'DB_CLOSE_DELAY=-1', I > would think the problem was the data cache being dropped across the > multiple connections. That or somehow the commit call on the connection is > not being handled properly. > > > -- You received this message because you are subscribed to the Google Groups "H2 Database" group. To view this discussion on the web visit https://groups.google.com/d/msg/h2-database/-/m--vFEGr3P4J. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/h2-database?hl=en.
