blktap kernel crash

[Frederic]

Hello,

We are running xen-4.1.2-2ubuntu2 on Ubuntu 12.04 amd64.
Our setup:
- 1 san server
- 1 ubuntu xen dom0 with iscsi/multipath configured & connection to our san.
We're working with raw disk images & those are stored on the san.
The virtual disks are used in the xen.cfg with "tap2:tapdisk:aio:"

**Problem:
When there is high IO (read/write) on vm's, the dom0-kernel crashes with following output:

123249.173995] BUG: soft lockup - CPU#3 stuck for 22s! [swapper/3:0]
[123249.173999] BUG: soft lockup - CPU#1 stuck for 22s! [swapper/1:0]
[123249.174275] Stack:
[123249.174287] Call Trace:
[123249.174288]
[123249.174377] Stack:
[123249.174392] Call Trace:
[123249.174393]
[123249.174974]
[123249.174997] Code: cc 51 41 53 b8 1c 00 00 00 0f 05 41 5b 59 c3 cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc 51 41 53 b8 1d 00 00 00 0f 05 <41> 5b 59 c3 cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc
[123249.175292] [] nf_iterate+0x85/0xc0
[123249.175403] Code: [] ? __xen_evtchn_do_upcall+0x207/0x250

**Used packages/versions:
linux-image-3.2.0-24-generic 3.2.0-24.39
blktap-dkms 2.0.91-1

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Andrea Luder (andrealuder)]

when I install UBUNTU 64bit distro =>16.04 on a Vmachine whit more than 1 processor
I got an unstable machine that still frozen ramdoly after few second.
watchdog BUG: soft lockup - CPU#3 stuck for 22s! [swapper/3:0]
cannot make diags and post logs because
don't work switch in tty alt+ctrl+Fx

distro 14.04 work fine

[Andrea Luder (andrealuder)]

when I install UBUNTU 64bit distro =>16.04 on a Vmachine whit more than 1 processor
I got an unstable machine that still frozen ramdoly after few second.
watchdog BUG: soft lockup - CPU#3 stuck for 22s! [swapper/3:0]
cannot make diags and post logs

distro 14.04 work fine