How to take down a system to the point of requiring a newfs with
one line of C (userland)
Kris Kennaway
kris at FreeBSD.org
Mon Feb 18 20:23:53 UTC 2008
Wesley Shields wrote:
> On Mon, Feb 18, 2008 at 07:21:55PM +0100, Volker wrote:
>> On 02/18/08 19:04, Wesley Shields wrote:
>>> On Mon, Feb 18, 2008 at 09:25:29AM -0800, Peter Sanchez wrote:
>>>> On Feb 18, 2008, at 7:07 AM, Wesley Shields wrote:
>>>>> I tried this using /tmp/ as argv[1] and it didn't crash a 6.2 machine or
>>>>> a -current from a few weeks ago. Maybe the number of files has to be
>>>>> increased? I bumped it up to 100000 and tried on a 6.2 machine, but I
>>>>> ran out of inodes before I could induce a crash. :)
>>>>>
>>>>> Maybe I'm doing something wrong?
>>>> I believe the panic doesn't occur until boot. Did you reboot the box after
>>>> writing the files to /tmp?
>>>>
>>>> Peter
>>> I did on a 6.2 machine with 10000 files in /tmp. I can reboot the
>>> -current machine later tonight if you think it will make a difference.
>> According to the problem report, it should panic while mounting the fs.
>> umount and re-mount /tmp and see, if you can make it panic (a reboot
>> shouldn't be necessary here).
>
> I did exactly that and it did not panic on both a 6.2 and -current
> machine.
>
> Just to be sure, I did reboot a 6.2 machine with 10000 0-byte files in
> /tmp and it didn't panic.
I'm also unable to reproduce this on an 8.0 machine. The original
poster needs to follow up with the panic backtrace so we can attempt to
understand his problem.
Kris
More information about the freebsd-security
mailing list