Let PCI device recover:
for i in `seq 1 100`; do echo 1 > /sys/bus/pci/devices/0000\:00\:00.0/recover; echo $i;sleep 1;done
On the 32nd recovery I got the kernel panic.
Retried that and on the 19th recovery the panic occured.
Updated system to 4.4.0-22.38 and performed the above steps. Problem did not occur after 2x 100 recoveries. Fix works. Waiting for integration into xenial stream.
@Canonical: Please integrate into xenial and yaketty stream.
------- Comment From <email address hidden> 2016-04-29 05:45 EDT-------
At first I reproduced this on kernel 4.4.0-15:
Creating workload:
cat /dev/urandom |gzip -9 > /dev/null &
cat /dev/urandom |gzip -9 > /dev/null &
ping 10.100.86.49
Let PCI device recover: pci/devices/ 0000\:00\ :00.0/recover; echo $i;sleep 1;done
for i in `seq 1 100`; do echo 1 > /sys/bus/
On the 32nd recovery I got the kernel panic.
Retried that and on the 19th recovery the panic occured.
Updated system to 4.4.0-22.38 and performed the above steps. Problem did not occur after 2x 100 recoveries. Fix works. Waiting for integration into xenial stream.
@Canonical: Please integrate into xenial and yaketty stream.