On Wednesday 05 December 2001 18:35, Oskar wrote: > Yes, but the interval is short enough that I don't think it matters. I > really don't think it's worth the complication to fix it. The idea of purposely ignoring race conditions rubs me the wrong way as a matter of principle.
Moreover, I think it might matter. Under moderate load you will get spurious timeouts as messages get "lost" behind messages with large trailing fields. I have seen something like this in my logs. Though I can't say for sure that this is the cause. > I strongly believe that you are continuing the grasp at straws with > this stuff. What stuff? Ignoring bugs doesn't make them go away. --gj > On Wed, Dec 05, 2001 at 04:27:24PM -0500, Gianni Johansson wrote: > > Hi Oskar, Tavin: > > > > Is OpenConnectionManager.findFreeConnection() supposed to remove the > > ConnectionHandler it returns from the table? It doesn't in the current > > code. > > > > If not, isn't there a race condition where two separate threads could > > call OpenConnectionManager.getConnection() and get the same "idle" > > ConnectionHandler? i.e. first call gets an idle handler, second call gets > > the same handler before the first can send a message on it. > > > > --gj > > > > -- > > Freesites > > (0.3) freenet:MSK at SSK@enI8YFo3gj8UVh-Au0HpKMftf6QQAgE/homepage// > > (0.4) freenet:SSK at npfV5XQijFkF6sXZvuO0o~kG4wEPAgM/homepage// > > > > _______________________________________________ > > Devl mailing list > > Devl at freenetproject.org > > http://lists.freenetproject.org/mailman/listinfo/devl -- Freesites (0.3) freenet:MSK at SSK@enI8YFo3gj8UVh-Au0HpKMftf6QQAgE/homepage// (0.4) freenet:SSK at npfV5XQijFkF6sXZvuO0o~kG4wEPAgM/homepage// _______________________________________________ Devl mailing list Devl at freenetproject.org http://lists.freenetproject.org/mailman/listinfo/devl
