[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [xmlblaster-devel] Socket test, was LOCAL not working?

On Wed, 2003-09-24 at 22:56, Marcel Ruff wrote:
> Peter Antman wrote:
> > By the way, debugging Jacorb is really hard. Including some debug stuff
> > makes the bug disappear or become more seldom.
> > 
> > I do have some traces which points in the direction that the waiting
> > thread will be awakened without any reply being received. Which part of
> > Jacorb that does this, or why no reply ever get received for these cases
> > I have not been able to find out.
> Peter,
> could you please try the SOCKET protocol
> if there are dead locks as well?

Hi, I just did that. During almost one and an half hour I could see no
signs of any lookups (the corba one does have contention really early,
and locks up eventually). However, finally this happened:

junit.framework.AssertionFailedError: Max number of new threads reached
516 number of threads created since first round: XmlBlaster is leaking
huge numbers of threads. Happened in round 330

As far as I have been able to see, it initially handles the thread count
well. The threads it creates are reclaimed after one or two rounds. But
eventually the number of threads created before they are reclaimed
grows, until it reached the set maximum.

> thanks
> Marcel
> -- 
> http://www.xmlBlaster.org
Peter Antman	Chief Technology Officer, Development
Technology in Media, Box 34105 100 26 Stockholm
WWW: http://www.tim.se	WWW: http://www.backsource.org
Email: pra at tim.se	 
Phone: +46-(0)8-506 381 11 Mobile: +46-(0)704 20 58 11