On Mon, Dec 7, 2009 at 11:20 AM, Alberto Treviño <[email protected]> wrote: > If you could trap SIGKILL, then it wouldn't work right and you could write > unkillable processes, much like Windows. We definitely don't want to do be > like Windows.
To be fair, you cannot write an "unkillable" process in Windows. There are circumstances, the likes of which you can intentionally replicate if you want, but which doing so serves no purpose, where a process appears to hang around after being terminated (i.e., it still shows up in Task Manager) but this is usually just the result of a kernel-mode portion of a process (device driver, etc) that is waiting on an IO request (or similar) and not terminating the way it should. When you terminate a process (via the TerminateProcess API) the user-mode portion is really torn down and gone. And, still being fair, the exact same kind of thing happens in Linux. If a program is waiting on some kind of interrupt or IO request, it's possible to send it SIGKILL have nothing seem to happen. It may also happen at other times, but I've never bothered to investigate it thoroughly when I've encountered it. In both cases if you have appropriate rights on the system you can usually attach a debugger to the process and determine the real culprit. Nick -------------------- BYU Unix Users Group http://uug.byu.edu/ The opinions expressed in this message are the responsibility of their author. They are not endorsed by BYU, the BYU CS Department or BYU-UUG. ___________________________________________________________________ List Info (unsubscribe here): http://uug.byu.edu/mailman/listinfo/uug-list
