A colleague at work can pretty reliably hang his XEmacs (21.5b28, Solaris 10) by double-clicking his right mouse button with this definition in place: (if XEmacs (progn (global-unset-key 'button3) (define-key global-map [button3 button3up button3] 'x-mouse-kill) (global-set-key 'button2 'x-yank-clipboard-selection) (global-set-key "\C-ce" 'cvs-examine) ) ) What he does is position the mouse with a button1 click, then double-click button3 to invoke x-mouse-kill. Top shows that XEmacs and Xorg are both busy. Running pstack against the XEmacs process suggests they are busy chatting with one another. I've tried it a few times on my machine but can't reproduce the problem. When he tries it he almost always provokes the problem. We are both running Solaris 10 on Dell hardware (dual-core single-CPU P4s running at 3.2 or 3.4GHz). I have a Dell keyboard and mouse. He has a Sun keyboard and mouse, all USB-connected. xdpyinfo shows this for my setup: version number: 11.0 vendor string: Sun Microsystems, Inc. vendor release number: 60900000 while this for his: version number: 11.0 vendor string: Sun Microsystems, Inc. vendor release number: 70200000 Both machines have dual-monitor single (X) screen setups, 3200x1200 resolution with similar/identical visuals. Does this ring a bell with anyone? I can see if the admins can back-rev his X server software, but I suspect they'll be disinclined to do that. Any other way to debug this problem? -- Skip Montanaro - skip(a)pobox.com - http://www.webfast.com/~skip/ "Be different, express yourself like everyone else." _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta