Launch instance got libvirtError for qemu unsupported IDE bus in AARCH64

[Kevin Zhao]

Description
===========
After setup the nova development environment with devstack in aarch64 machine ,use the glance upload the image ,then use nova to launch the instance.Launching failed with the error "libvirtError: unsupported configuration: IDE controllers are unsupported for this QEMU binary or machine type".

Steps to reproduce
==================
1.Using devstack to deploy openstack. Using default local.conf.

2.Upload the aarch64 image with glance.
$ source ~/devstack/openrc admin admin
$ glance image-create --name image-arm64.img --disk-format qcow2 --container-format bare --visibility public --file images/image-arm64-wily.qcow2 --progress
$ glance image-create --name image-arm64.vmlinuz --disk-format aki --container-format aki --visibility public --file images/image-arm64-wily.vmlinuz --progress
$ glance image-create --name image-arm64.initrd --disk-format ari --container-format ari --visibility public --file images/image-arm64-wily.initrd --progress
$ IMAGE_UUID=$(glance image-list | grep image-arm64.img | awk '{ print $2 }')
$ IMAGE_KERNEL_UUID=$(glance image-list | grep image-arm64.vmlinuz | awk '{ print $2 }')
$ IMAGE_INITRD_UUID=$(glance image-list | grep image-arm64.initrd | awk '{ print $2 }')
$ glance image-update --kernel-id ${IMAGE_KERNEL_UUID} --ramdisk-id ${IMAGE_INITRD_UUID} ${IMAGE_UUID}

3.nova add keypair
$ nova keypair-add default --pub-key ~/.ssh/id_rsa.pub

4.Launch the instance:
$ image=$(nova image-list | egrep "image-arm64.img"'[^-]' | awk '{ print $2 }')
$ nova boot --flavor m1.medium --image ${image} --key-name default test-arm64

5.screen -x and select the n-cpu session to see the output.
Then will got the error.

Expected result
===============
After spawningn the instance, use :
$ nova list
We can see the instance is active.

Actual result
=============
Got the error:
libvirtError: unsupported configuration: IDE controllers are unsupported for this QEMU binary or machine type

We can see the detailed information：
 ERROR nova.compute.manager [req-75325207-6c1b-481d-b188-a66c0a64eb89 admin admin] [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] Instance failed to spawn
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] Traceback (most recent call last):
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/opt/stack/nova/nova/compute/manager.py", line 2041, in _build_resources
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] yield resources
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/opt/stack/nova/nova/compute/manager.py", line 1887, in _build_and_run_instance
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] block_device_info=block_device_info)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/opt/stack/nova/nova/virt/libvirt/driver.py", line 2569, in spawn
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] block_device_info=block_device_info)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/opt/stack/nova/nova/virt/libvirt/driver.py", line 4713, in _create_domain_and_network
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] xml, pause=pause, power_on=power_on)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/opt/stack/nova/nova/virt/libvirt/driver.py", line 4644, in _create_domain
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] guest.launch(pause=pause)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/opt/stack/nova/nova/virt/libvirt/guest.py", line 142, in launch
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] self._encoded_xml, errors='ignore')
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/usr/local/lib/python2.7/dist-packages/oslo_utils/excutils.py", line 221, in __exit__
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] self.force_reraise()
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/usr/local/lib/python2.7/dist-packages/oslo_utils/excutils.py", line 197, in force_reraise
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] six.reraise(self.type_, self.value, self.tb)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/opt/stack/nova/nova/virt/libvirt/guest.py", line 137, in launch
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] return self._domain.createWithFlags(flags)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/usr/local/lib/python2.7/dist-packages/eventlet/tpool.py", line 186, in doit
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] result = proxy_call(self._autowrap, f, *args, **kwargs)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/usr/local/lib/python2.7/dist-packages/eventlet/tpool.py", line 144, in proxy_call
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] rv = execute(f, *args, **kwargs)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/usr/local/lib/python2.7/dist-packages/eventlet/tpool.py", line 125, in execute
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] six.reraise(c, e, tb)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/usr/local/lib/python2.7/dist-packages/eventlet/tpool.py", line 83, in tworker
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] rv = meth(*args, **kwargs)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] File "/usr/local/lib/python2.7/dist-packages/libvirt.py", line 1065, in createWithFlags
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] if ret == -1: raise libvirtError ('virDomainCreateWithFlags() failed', dom=self)
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] libvirtError: unsupported configuration: IDE controllers are unsupported for this QEMU binary or machine type
 TRACE nova.compute.manager [instance: 188aa5bc-173c-46ec-b872-6bacb512911e]
 INFO nova.compute.manager [req-75325207-6c1b-481d-b188-a66c0a64eb89 admin admin] [instance: 188aa5bc-173c-46ec-b872-6bacb512911e] Terminating instance

Environment
===========
1. Exact version of OpenStack you are running. See the following
  list for all releases: http://docs.openstack.org/releases/
   Nova development, commit code: 9a05d38f48ef0f630c5e49e332075b273cee38b9

2. Which hypervisor did you use?
    Libvirt+KVM
    $ kvm --version
    QEMU emulator version 2.5.0 (Debian 1:2.5+dfsg-5ubuntu10.1), Copyright (c) 2003-2008 Fabrice Bellard
    $ libvirtd --version
    libvirtd (libvirt) 1.3.1

2. Which storage type did you use?
   In the host file system,all in one physics machine.q
stack@u202154:/opt/stack/nova$ df -hl
Filesystem Size Used Avail Use% Mounted on
udev 7.8G 0 7.8G 0% /dev
tmpfs 1.6G 61M 1.6G 4% /run
/dev/sda2 917G 41G 830G 5% /
tmpfs 7.9G 0 7.9G 0% /dev/shm
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 7.9G 0 7.9G 0% /sys/fs/cgroup
/dev/sda1 511M 888K 511M 1% /boot/efi
cgmfs 100K 0 100K 0% /run/cgmanager/fs
tmpfs 1.6G 0 1.6G 0% /run/user/1002
tmpfs 1.6G 0 1.6G 0% /run/user/1000
tmpfs 1.6G 0 1.6G 0% /run/user/0

3. Which networking type did you use?
   nova-network

4. Environment information:
   Architecture : AARCH64
   OS: Ubuntu 16.04

Detailed log info is in the accessory.
The guest xml is also in the log info.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/323244

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/323244
Committed: https://git.openstack.org/cgit/openstack/nova/commit/?id=36998c05d5b3c23a6b35bd0620d8a0f5f7e5d80c
Submitter: Jenkins
Branch: master

commit 36998c05d5b3c23a6b35bd0620d8a0f5f7e5d80c
Author: Kevin Zhao <email address hidden>
Date: Tue May 31 02:54:40 2016 +0000

    Specify the default cdrom type "scsi" for AARCH64.

    In AARCH64,ide is not the proper type for cdrom device just like
    in x86_64.So change this to scsi in case of qemu error. Also add
    a test for this.

    Close-Bug: #1585893

    Change-Id: Ia1cfde9f96fa54de0b352a7e6e903898b1eb7476
    Signed-off-by: Kevin Zhao <email address hidden>

[Augustina Ragwitz (auggy)]

Since the Mitaka cycle we use the direct release model, which means this should be Fix Released.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/mitaka
Review: https://review.openstack.org/387659

[OpenStack Infra (hudson-openstack)]

Change abandoned by Dirk Mueller (<email address hidden>) on branch: stable/mitaka
Review: https://review.openstack.org/387659