When ibus is running, starting virt-manager and double-clicking a machine hangs the process.  strace shows it repeatedly sched_yield()ing.  This seems to have something to do with the interaction between dbus and virt-manager forking off a new process (so it runs fine with --no-fork).  The reason that enabling ibus seems to trigger this is that ibus tries talks to its own dbus bus, which is where it seems to hang in the strace.  The full output of strace -f python src/virt-manager.py is at http://ricky.fedorapeople.org/virt-manager.strace.out.

Moving the forking code (in virt-manager.py):

    # Now we've got basic environment up & running we can fork
    if not options.nofork and not options.debug:
        drop_tty()
        drop_stdio()

        # Ignore SIGHUP, otherwise a serial console closing drops the whole app
        signal.signal(signal.SIGHUP, signal.SIG_IGN)

anywhere above the

    config = virtManager.config.vmmConfig(appname, appversion, ui_dir)

line seems to avoid the hang.  Unfortunately I haven't gotten much further in trying to analyze this bug, as I'm not very familiar with dbus or gtk.

I've reproduced this with virt-manager-0.9.1-2.fc15.noarch as well as current code in git master on Fedora 15.