libvirt: Wrap un-proxied listDevices() and listAllDevices()

This is similar to change I668643c836d46a25df46d4c99a973af5e50a39db
where the objects returned in a list from a libvirt call were not
tpool.Proxy wrapped. Because the objects are not wrapped, calling
methods on them such as listCaps() can block all other greenthreads
and can cause nova-compute to freeze for hours in certain scenarios.

This adds the same wrapping to libvirt calls which return lists of
virNodeDevice.

Closes-Bug: #2091033

Change-Id: I60d6f04d374e9ede5895a43b7a75e955b0fea3c5
This commit is contained in:
melanie witt
2024-10-18 02:54:02 +00:00
committed by Stephen Finucane
parent a67ab85678
commit f304b9eaad
3 changed files with 60 additions and 2 deletions

View File

@@ -2245,6 +2245,48 @@ class LibvirtTpoolProxyTestCase(test.NoDBTestCase):
self.assertIsInstance(domain, tpool.Proxy)
self.assertIn(domain.UUIDString(), (uuids.vm1, uuids.vm2))
def _add_fake_host_devices(self):
self.conn._obj.pci_info = fakelibvirt.HostPCIDevicesInfo(
num_pci=1, num_pfs=2, num_vfs=2, num_mdevcap=3)
mdevs = {
'mdev_4b20d080_1b54_4048_85b3_a6a62d165c01':
fakelibvirt.FakeMdevDevice(
dev_name='mdev_4b20d080_1b54_4048_85b3_a6a62d165c01',
type_id=fakelibvirt.NVIDIA_11_VGPU_TYPE,
parent=fakelibvirt.MDEVCAP_DEV1_PCI_ADDR),
}
self.conn._obj.mdev_info = fakelibvirt.HostMdevDevicesInfo(
devices=mdevs)
def test_tpool_list_all_devices(self):
self._add_fake_host_devices()
devs = self.host.list_all_devices(
fakelibvirt.VIR_CONNECT_LIST_NODE_DEVICES_CAP_PCI_DEV)
self.assertEqual(8, len(devs))
for dev in devs:
self.assertIsInstance(dev, tpool.Proxy)
def test_tpool_list_pci_devices(self):
self._add_fake_host_devices()
devs = self.host.list_pci_devices()
self.assertEqual(8, len(devs))
for dev in devs:
self.assertIsInstance(dev, tpool.Proxy)
def test_tpool_list_mdev_capable_devices(self):
self._add_fake_host_devices()
devs = self.host.list_mdev_capable_devices()
self.assertEqual(3, len(devs))
for dev in devs:
self.assertIsInstance(dev, tpool.Proxy)
def test_tpool_list_mediated_devices(self):
self._add_fake_host_devices()
devs = self.host.list_mediated_devices()
self.assertEqual(1, len(devs))
for dev in devs:
self.assertIsInstance(dev, tpool.Proxy)
class LoadersTestCase(test.NoDBTestCase):

View File

@@ -1647,7 +1647,9 @@ class Host(object):
:returns: a list of virNodeDevice instance
"""
try:
return self.get_connection().listDevices(cap, flags)
devs = [self._wrap_libvirt_proxy(dev)
for dev in self.get_connection().listDevices(cap, flags)]
return devs
except libvirt.libvirtError as ex:
error_code = ex.get_error_code()
if error_code == libvirt.VIR_ERR_NO_SUPPORT:
@@ -1667,7 +1669,10 @@ class Host(object):
:returns: a list of virNodeDevice instances.
"""
try:
return self.get_connection().listAllDevices(flags) or []
alldevs = [
self._wrap_libvirt_proxy(dev)
for dev in self.get_connection().listAllDevices(flags)] or []
return alldevs
except libvirt.libvirtError as ex:
LOG.warning(ex)
return []

View File

@@ -0,0 +1,11 @@
fixes:
- |
`Bug #2091033`_: Fixed calls to libvirt ``listDevices()`` and
``listAllDevices()`` from potentially blocking all other greenthreads
in ``nova-compute``. Under certain circumstances, it was possible for
the ``nova-compute`` service to freeze with all other greenthreads
blocked and unable to perform any other activities including logging.
This issue has been fixed by wrapping the libvirt ``listDevices()``
and ``listAllDevices()`` calls with ``eventlet.tpool.Proxy``.
.. _Bug #2091033: https://bugs.launchpad.net/nova/+bug/2091033