Fwd: SOLVED: 1.4.1 Swing Apps Hanging

From: Andrus Adamchik (andru..bjectstyle.org)
Date: Sat Mar 15 2003 - 19:04:06 EST

  • Next message: Craig Miskell: "Re: DataContext.clearFlattenedUpdateQueries vs. ContextCommit"

    For those running CayenneModeler / JDK 1.4.1 on Mac OS X. There is
    indeed a bug (it was submitted to Apple already). I just received a
    confirmation of that (see below). Oh well, we will have to wait some
    more.

    Andrus

    Begin forwarded message:

    > From: Vadim Penzin
    > Date: Sat Mar 15, 2003 6:40:23 PM US/Eastern
    > To: java-de..ists.apple.com
    > Subject: SOLVED: 1.4.1 Swing Apps Hanging
    >
    > One word: UFS.
    >
    > After talking with a friend whose computer did not manifestate the
    > bug, I realised that there is probably only one significant
    > difference between our machines: he uses HFS+ and I installed mine on
    > UFS.
    >
    > I took the following steps to verify that:
    >
    > 1. I downloaded Mac OS X 10.2.4 Combined Update and Java 1.4.1 and
    > burned them on a CD.
    > 2. I made a clean installation of Mac OS 10.2.1 from the CD supplied
    > with my computer (Power Book G4 867 MHz/512 MB/60 GB) using *UFS*
    > partition spanning the entire drive.
    > 3. I ignored notifications from Software Update.
    > 4. I installed Mac OS X 10.2.4 Combined Update from the CD.
    > 5. I installed Java 1.4.1 from the CD.
    >
    > The bug was still there -- JVM deadlocked when I launched my test code:
    >
    > import javax.swing.JFrame;
    > import java.awt.Dimension;
    >
    > class Foo extends JFrame
    > {
    > public Foo ()
    > {
    > super ( "Foo" );
    > setSize ( 100, 100 );
    > setVisible ( true );
    > }
    >
    > public static void main ( String [] args )
    > {
    > new Foo ();
    > }
    > }
    >
    > Then I repeated everything (points 2-5 above) anew but with HFS+
    > instead of UFS in point 2 -- my test code worked.
    >
    > As a side note, I felt my system's performance improved with HFS+.
    >
    > I admit, migrating to HFS+ from UFS is not actually a solution and, of
    > course, it does not solve the bug -- now it is Apple's turn.
    >
    >
    >



    This archive was generated by hypermail 2.0.0 : Sat Mar 15 2003 - 19:07:33 EST