Comment 6 for bug 1825413
Revision history for this message
| Seth Forshee (sforshee) wrote Re: [Bug 1825413] Re: mdadm, mkfs, other io commands hang, stuck task, bad rip | #6 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
0e1f616
(Get the code!)
On Mon, Apr 22, 2019 at 08:20:30PM -0000, Ryan Harper wrote:
> Hi Seth,
>
> notice only one of the stack tracks have the floppy, the mdadm one does
> not. I've also recreated this on a qemu q35 machine type which does not
> include the floppy device.
I looked at the madm stack trace, and I was assuming that it occurred
during the same boot as the other one. I suppose I shouldn't have, but
it would make sense. mdadm appears to be blocked on a block device's
bd_mutex, so the cause is probably something else that has hung while
holding that mutex. The mkfs stack trace would be just the kind of thing
to create the mdadm stack trace, if it were trying to open the same
device.
So, were those traces from a single boot or from different boots?
Can you provide stack trace(s) from the q35 machine type for comparison?
Thanks!