Sounds like the bug is the md driver cloning "md10ec" which it shouldn't
do. This bug must naturally be in md_clone(), but I don't have the
minutes right now to hunt it down.
Should be quite simple to nail, it's just some string handling code.
In message <[EMAIL PROTECTED]>, Chad David writes:
>I posted a bug report and patch in kern/29104 and Dima Dorfman also mentioned
>this in July/August, but it still has not been resolved. The method of
>triggering it that I detailed in my bug report no longer seems to work,
>but I've managed to create another one.
># mdconfig -a -t swap -s 32m -u 10
># disklabel -r -w md10 auto
># disklabel -e md10e (copy c to e and set type to 4.2BSD)
># ls -l /dev/md10*
>crw-r----- 1 root operator 95, 0x00010052 Oct 19 01:50 /dev/md10
>crw-r----- 1 root operator 95, 82 Oct 19 02:00 /dev/md10c
>crw-r----- 1 root operator 95, 84 Oct 19 01:55 /dev/md10ec
>crw------- 1 root wheel 95, 0xffff00ff Oct 19 01:50 /dev/mdctl
>Now if you access (open(2)) md10e the system will kill itself when you
>run mdconfig -d. If I am doing something evil I'd like to know, but either
>way the kernel should not enter an endless loop!
>I have not had time to determine why disklabel -e md10e will kill it, but
>disklabel -e /dev/md10e will not. I think it has to do with how disklabel
>appends the 'c'.
>The patch that I provided in 29104 is probably not the correct solution
>(I like Dima's default better), and may not even be "fixing" the problem
>in the correct place. Any advice would be very welcome.
>Chad David [EMAIL PROTECTED]
>ACNS Inc. Calgary, Alberta Canada
>To Unsubscribe: send mail to [EMAIL PROTECTED]
>with "unsubscribe freebsd-current" in the body of the message
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
[EMAIL PROTECTED] | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message