Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo


#115 hang on workspace load after crash

Bug (46)
Michal Konečný

I observed that every time when I open Eclipse with a workspace that contains several Haskell projects (using eclipseFP 1.110.0)
after a system crash, it stops for at least 1-2 minutes on the splash stuck on "Loading Workspace" before opening
the workspace.

The last time it happened, the splash screen keeps getting stuck on at this state for several hours and then I gave up.
The workspace log from around the time when I tried start up says:

!ENTRY org.eclipse.osgi 2 0 2010-05-24 10:31:26.182
!MESSAGE While loading class "net.sf.eclipsefp.haskell.ui.internal.editors.haskell.text.LiterateHaskellPartitionScanner", thread "Thread[main,6,main]"
timed out waiting (5000ms) for thread "Thread[Worker-5,5,main]" to finish starting bundle "net.sf.eclipsefp.haskell.ui_1.110.0 [905]". To avoid deadl
ock, thread "Thread[main,6,main]" is proceeding but "net.sf.eclipsefp.haskell.ui.internal.editors.haskell.text.LiterateHaskellPartitionScanner" may no
t be fully initialized.
org.osgi.framework.BundleException: State change in progress for bundle "reference:file:plugins/net.sf.eclipsefp.haskell.ui_1.110.0.jar" by thread "Wo
at org.eclipse.osgi.framework.internal.core.AbstractBundle.beginStateChange(AbstractBundle.java:1073)
(followed by a long Java exception trace)

I managed to recover from the situation by first opening the projects in another workspace, which failed with strange results
and then reopening the old workspace, this time with no delay.


  • I submitted this bug using Google authentication that SourceForge offered me not knowing they would not show my identity properly... my name is Michal Konecny and email mikkonecny@gmail.com.

  • JP Moresmau
    JP Moresmau

    The long Java exception trace could be useful to know where the deadlock was. Can you reproduce?

  • I have not had this problem for a long time - I think since I upgraded to Helios.
    I could not reproduce it even by trying hard to kill Eclipse in the most brutal ways.
    I also sifted all logs still kept in my workspaces and unfortunately all traces of the bug
    have been rotated away.