fdavis112@juno.com wrote:
> Wouldn't in be nice to have a system that once a kernel is built,
> implement the new kernel without a software or hardware reboot?
Yes, it would be very nice. I plan to work on such a mechanism later
this year (unless somebody else is faster :).
What I basically want to do is to use a Linux kernel as the
second-stage boot loader. This (supposedly stable) kernel can then
load the "real" kernel (and possibly initrd) from wherever they may
be hiding (e.g. RAID, tftp through an IPsec tunnel, you name it).
Then it passes any useful information it may have gathered to far
(e.g. data from the real-mode BIOS, SCSI bus scan results, etc.) to
the new kernel and boots it.
It could also be of some limited use for accelerating kernel
development. Of course, if you crash the system, it will be as dead
as it is now.
> And, if there were any run-time problems with the new kernel,
> the system would revert back to the previously runnable kernel.
This would be a lot more difficult. What you can do today, with
LILO, is to specify a fallback image. Then add a watchdog device,
and if your new kernel gets in trouble, the watchdog resets the
system, and you boot the fallback kernel.
- Werner
-- _________________________________________________________________________ / Werner Almesberger, ICA, EPFL, CH werner.almesberger@ica.epfl.ch / /_IN_N_032__Tel_+41_21_693_6621__Fax_+41_21_693_6610_____________________/- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Jan 15 2000 - 21:00:12 EST