H. Hartzer wrote:

> bsd# dd if=/dev/null of=1 bs=1m seek=8000000; (ulimit -t 1; cmp 1 1)
> 0+0 records in
> 0+0 records out
> 0 bytes transferred in 0.000 secs (0 bytes/sec)
> Killed
> bsd#
> 
> It took a while to get to Killed.
> 
> There's no message in dmesg.
> 
> Does it definitely crash for you, or just unresponsive?

Hi Henrich,

The way I trigger it is different. With the sparse file created and
in the current directory, I execute a grep - the bug seems to be
triggered after initiating a ^C to the grep. CPU will climb to max,
any open SSH sessions time out and the console is unresponsive to
any input. I suppose I didn't wait long enough but the one time I
was able to trigger ddb it noted a kernel: double fault trap so I
assume the system is halted at that point. It required a poweroff
of the VM to recover.

Either way, a non-root user should not be able to do this!

I forgot to mention: I was testing this on amd64 VMware ESXi.

I'm curious what would happen on other architectures like SPARC.

Regards
Lloyd


Reply via email to