> If the init process exits, the kernel panics. Therefore no normal > distro will ever have an init that returns to its caller. (also, if you > didn't exec into the new init, then it wouldn't be process 0, and could > cause you problems)
You mean process 1. ;) Actually, mainstream "init" processes usually detect when they are not run as process 1 and do something else entirely in this case (i.e. send a command to process 1 via /dev/initctl). At least System V init and Upstart do this. So, Michael *is* properly executing into his main init - else his system likely would not boot at all. [ Side note : panicking when process 1 exits is a *very* silly thing for the kernel to do. Process 1 exiting, instead of being forbidden, should mean the end of the machine's life cycle, and the kernel should either halt, reboot, or even kexec something else, depending on process 1's exit code. But that is totally off-topic. ] Some startup scripts fail to properly cleanup the boot environment. Ubuntu, up to at least 10.04, keeps two instances of tmpfs mounts that are only used during the early boot stages. But nobody cares because nobody's using Ubuntu on a machine with scarce memory resources anyway. For all intents and purposes, once the full system has booted, no trace of the initial boot environment remains. > Most likely you want to alter the shutdown scripts of your distro. Yes. The initial environment only exists to provide your box with the minimum requirements to be able to boot the "full" system. Once the full system has booted, there is no need to "go back" to that initial environment, ever : the full system provides everything that is necessary to work on the machine. If you want to do something at shutdown time, the proper place to do it is the full system's shutdown scripts. And none of this has the slightest thing to do with busybox. ;) -- Laurent _______________________________________________ busybox mailing list [email protected] http://lists.busybox.net/mailman/listinfo/busybox
