Messages in this thread |  | | | From | Sprag Johnson <> | | Subject | (dumb?) idea for kernel oops... | | Date | Mon, 5 Oct 1998 22:00:12 -0500 (EST) |
| |
I was thinking about when the kernel hangs or otherwise dies with an oops...what's to keep it from saving the crash info in a known memory location and restarting the kernel from scratch? Its dead anyway, so its not like you're going to lose any information...but you might be able to save some.
When the kernel is starting up, it could check the "i'm rebooting from a hang" flag and do some cleanups (possibly). Things like adding crash messages to the log would be good, as well as possibly saving data in the caches. If the kernel doesn't come all the way back up, it would have to die, but if it did come up, the flag could be cleared, and the system would be available for inspection.
Dumb idea? It would make life easier for those of us who administer machines from a distance (Don't need to be there to reboot), and would probably be a boost to Linux's availablity rating...
Brian Wheeler bdwheele@indiana.edu
- 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/
|  |