On 17/08/2012 01:19, Jonathan Tripathy wrote:
On 17/08/2012 00:34, Kent Overstreet wrote:
On Thu, Aug 16, 2012 at 06:51:15PM +1000, Joseph Glanville wrote:
Hi list, Kent,
I caused a crash with the current 3.2 branch under Xen dom0.
By the look of it the swap LV caused the crash?
Argh. Strange bug.
I wonder if this is related to a bug someone on lkml noticed the other
day in my change to the dm code... I see you've got dm_multipath linked
in. That would cause memory corruption (underruning allocated memory),
which might cause this.
I'm going to try and reproduce it too, it might be something simpler.
How are things configured? I see raid1 in the call stack, but there must
be something else too because raid1 doesn't call bio_pair_split().
Just to let you guys know that I'm currently compiling the bcache-3.2
branch as I type this and will be testing bcache on a Xen Dom0 test
bed over the next few weeks. OS is Ubuntu 12.04.x86_64. Xen version
will be whatever is in the Ubuntu universe repo (4.1.2). Kent, thanks
for your commit within the last hour. Was this to fix Joseph's problem?
Thanks
--
I should also add that my backend spindle storage is a MD RAID10 array.
My to-be cache device is a MD RAID1 array of 2 SSD drives.
Cheers
--
To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
