Thanks so much Francois, I appreciate the effort you applied to validating
this.
Cheers,
Clinton
2010/5/14 François Schiettecatte
> Clinton
>
> Ok this is interesting, I was able to reliably replicate the problem, I
> even grabbed the source code and included in the source code tree of my
> app
Clinton
Ok this is interesting, I was able to reliably replicate the problem, I even
grabbed the source code and included in the source code tree of my application
and still got the issue, but... the server I am running this one was rebooted
due to a power failure and now the problem has gone a
Clinton
I am getting closer, I was able to make the issue go away by selectively
rolling back some of the changes that were made between the two version around
the handling of transaction and rollbacks. The problem is that I dont have
sure-fire test case for this, I have to run a crawler for 3-
Clinton
I understand, I am looking into it now, if I cant verify it before the weekend
is over then I would say that something else is going on in my config.
François
On May 8, 2010, at 9:41 AM, Clinton Begin wrote:
> Hi Francois,
>
> I can't recreate this problem. I've verified in about 10
Hi Francois,
I can't recreate this problem. I've verified in about 1000 unit tests with
different configurations that close is being called. The only known and
verified case of this was with the ManagedTransaction, not the
JdbcTransaction.
Perhaps you can put together a simple unit test with HS
Clinton
Thanks for pinging me.
I have not had a chance to look at it, and it is still an issue, before rushing
out and creating a jira ticket for it I wanted to look at the code diffs
between the two releases, I will try to get to it later today or this WE.
François
On May 7, 2010, at 10:33 A
Let me know if this is still a problem, and create a JIRA ticket if it is.
I should have some time this weekend to look into it.
Clinton
On Wed, May 5, 2010 at 9:16 PM, Clinton Begin wrote:
> Two things to do:
>
> * Create a JIRA ticket for this. We'll continue conversation and
> investigation
Two things to do:
* Create a JIRA ticket for this. We'll continue conversation and
investigation there.
* Mount the source and step through one example transaction. See if
you can verify that close isn't being called, and possibly why.
* Finally, out of curiosity, disable caching globally, and
Clinton
A little more information on this, I reverted back to 240 and the problem I
reported did not occurs after a 24 hour run. I then went to the current release
and the problem reappeared, everything else is the same, so there is an issue.
What it looks like is going on is that connections a
Then no, I don't believe there have been any changes related to
connections or transactions between those versions. You can have a
look at the change log if you like:
https://issues.apache.org/jira/secure/BrowseProject.jspa?id=10601&subset=-1
Or the subversion history as well.
Clinton
2010/5/1
I am using JDBC, with C3P0 as my data source.
Cheers
François
On May 1, 2010, at 4:14 PM, Clinton Begin wrote:
> The only code changes related to that were with managed transactions.
> Which transacion manager are you using? (JDBC or MANAGED)?
>
> Cheers,
> Clinton
>
> 2010/5/1 François Schie
The only code changes related to that were with managed transactions.
Which transacion manager are you using? (JDBC or MANAGED)?
Cheers,
Clinton
2010/5/1 François Schiettecatte :
> Hi
>
> I grabbed the latest, greatest release of 3.0 from :
>
>
> http://people.apache.org/builds/ibatis/ib
12 matches
Mail list logo
