In that case, it is an issue.
Can someone else also make comments about the impact of removing the
factory.destroy()?
I know the Oozie team at Y! made substantial investigations in the DB access
layer.
Regards,
Mohammad
On Monday, June 9, 2014 4:44 PM, bowen zhang
I don't think it will be wise to remove that line as factory.close() is
required to cleanup and release resources. Since the exception is during
shutdown and is just logged and does not affect the shutdown I don't think
you should be worrying about removing it.
On Mon, Jun 9, 2014 at 11:24 PM,
no, the exception will stop the shutdown. It will make oozie unable to shutdown.
On Tuesday, June 10, 2014 12:46 PM, Rohini Palaniswamy
rohini.adi...@gmail.com wrote:
I don't think it will be wise to remove that line as factory.close() is
required to cleanup and release resources. Since
Hi Mohammad,
I saw you wrote the line factory.close() in the destroy method of
JPAService.java. I've seen some issues of this line throwing
invalidStateException while there is active transaction happening,
particularly on Windows. Can you tell me how you make sure there is no
active
Bowen,
This has been done a long time back. I can't recall the full context.
However, it is better to make sure all transactions are completed before
closing it.
Is there a way to do that? This will be a safer approach.
Deleting the line looks little risky. We need to close the factory.
Mohammad, there is no code insuring all active transactions are closed.
On Monday, June 9, 2014 2:53 PM, Mohammad Islam misla...@yahoo.com.INVALID
wrote:
Bowen,
This has been done a long time back. I can't recall the full context.
However, it is better to make sure all transactions are