Bug 1318767 - live migration without shared storage fails in pre_live_migration after upgrade to 2015.1.2-18.2
Summary: live migration without shared storage fails in pre_live_migration after upgra...
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat OpenStack
Classification: Red Hat
Component: openstack-nova
Version: 5.0 (RHEL 7)
Hardware: Unspecified
OS: Unspecified
urgent
urgent
Target Milestone: ---
: 5.0 (RHEL 7)
Assignee: Eoghan Glynn
QA Contact: Prasanth Anbalagan
URL:
Whiteboard:
Depends On: 1318722
Blocks:
TreeView+ depends on / blocked
 
Reported: 2016-03-17 18:22 UTC by Lee Yarwood
Modified: 2019-10-10 11:35 UTC (History)
13 users (show)

Fixed In Version: openstack-nova-2014.1.5-30.el7ost
Doc Type: Bug Fix
Doc Text:
Clone Of: 1318722
Environment:
Last Closed: 2016-04-26 15:39:41 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
OpenStack gerrit 294205 0 None MERGED libvirt: Decode disk_info before use 2020-10-22 11:20:56 UTC
Red Hat Knowledge Base (Solution) 2202961 0 None None None 2016-03-17 18:22:11 UTC
Red Hat Product Errata RHBA-2016:0690 0 normal SHIPPED_LIVE openstack-nova bug fix advisory 2016-04-26 19:39:14 UTC

Description Lee Yarwood 2016-03-17 18:22:12 UTC
+++ This bug was initially created as a clone of Bug #1318722 +++

Description of problem:

After the upgrade to nova 2015.1.2-18.2 - CVE-2016-2140 fix, live migration fails without shared storage.

Version-Release number of selected component (if applicable):
* python-nova-2015.1.2-18.2.el7ost.noarch

How reproducible:
always

Steps to Reproduce:
1. configure live migration without shared storage
2. nova live-migration --block-migrate 8e972bd1-7e82-4868-9ac4-b80cc2eb098e osp7-compute
3. migration fails in pre_live_migration:

Actual results:
2016-03-17 11:02:45.192 5674 ERROR nova.compute.manager [req-da974863-f48d-4a20-8d95-133ddc39acda 4b730783a9af469b8168a8e8a58e510f 342bdc50ad5c415594975e762bdd8456 - - -] [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] Pre live migration failed at osp7-compute
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] Traceback (most recent call last):
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/compute/manager.py", line 5308, in _do_live_migration
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     block_migration, disk, dest, migrate_data)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/compute/rpcapi.py", line 627, in pre_live_migration
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     disk=disk, migrate_data=migrate_data)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/oslo_messaging/rpc/client.py", line 156, in call
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     retry=self.retry)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/oslo_messaging/transport.py", line 90, in _send
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     timeout=timeout, retry=retry)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/oslo_messaging/_drivers/amqpdriver.py", line 350, in send
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     retry=retry)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/oslo_messaging/_drivers/amqpdriver.py", line 341, in _send
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     raise result
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] TypeError: string indices must be integers
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] Traceback (most recent call last):
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/oslo_messaging/rpc/dispatcher.py", line 142, in _dispatch_and_reply
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     executor_callback))
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/oslo_messaging/rpc/dispatcher.py", line 186, in _dispatch
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     executor_callback)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/oslo_messaging/rpc/dispatcher.py", line 130, in _do_dispatch
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     result = func(ctxt, **new_args)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/compute/manager.py", line 6845, in pre_live_migration
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     disk, migrate_data=migrate_data)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/compute/manager.py", line 461, in decorated_function
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     return function(self, context, *args, **kwargs)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/exception.py", line 88, in wrapped
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     payload)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/oslo_utils/excutils.py", line 85, in __exit__
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     six.reraise(self.type_, self.value, self.tb)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/exception.py", line 71, in wrapped
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     return f(self, context, *args, **kw)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/compute/manager.py", line 369, in decorated_function
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     kwargs['instance'], e, sys.exc_info())
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/oslo_utils/excutils.py", line 85, in __exit__
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     six.reraise(self.type_, self.value, self.tb)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/compute/manager.py", line 357, in decorated_function
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     return function(self, context, *args, **kwargs)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/compute/manager.py", line 5272, in pre_live_migration
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     migrate_data)
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]   File "/usr/lib/python2.7/site-packages/nova/virt/libvirt/driver.py", line 6003, in pre_live_migration
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e]     image_file = os.path.basename(info['path'])
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] TypeError: string indices must be integers
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 
2016-03-17 11:02:45.192 5674 TRACE nova.compute.manager [instance: 8e972bd1-7e82-4868-9ac4-b80cc2eb098e] 


   5997             # Recreate the disk.info file and in doing so stop the
   5998             # imagebackend from recreating it incorrectly by inspecting the
   5999             # contents of each file when using the Raw backend.
   6000             if disk_info:
   6001                 image_disk_info = {}
   6002                 for info in disk_info:
-->   6003                     image_file = os.path.basename(info['path'])
   6004                     image_path = os.path.join(instance_dir, image_file)
   6005                     image_disk_info[image_path] = info['type']
   6006 


Expected results:
migration works

Additional info:

when downgrade to 2015.1.2-18 migration works.

--- Additional comment from Martin Schuppert on 2016-03-17 11:45:43 EDT ---

disk_info is:

2016-03-17 11:37:42.637 6497 DEBUG nova.virt.libvirt.driver [req-8c0157c2-adf7-41c4-bbf3-7117c96d5fa8 4b730783a9af469b8168a8e8a58e510f 342bdc50ad5c415594975e762bdd8456 - - -] disk_info: u'[{"disk_size": 1703936, "backing_file": "7aadcb85f689579f6b1cf9b7d21bfaed4212f42f", "virt_disk
_size": 1073741824, "path": "/var/lib/nova/instances/8e972bd1-7e82-4868-9ac4-b80cc2eb098e/disk", "type": "qcow2", "over_committed_disk_size": 1072037888}]' pre_live_migration /usr/lib/python2.7/site-packages/nova/virt/libvirt/driver.py:6000

--- Additional comment from Lee Yarwood on 2016-03-17 12:10:42 EDT ---

This is happening due to self.driver.get_instance_disk_info being called by the compute manager when block_migration is enabled :

nova/compute/manager.py

5293     def _do_live_migration(self, context, dest, instance, block_migration,
5294                            migrate_data):
5295         # Create a local copy since we'll be modifying the dictionary
5296         migrate_data = dict(migrate_data or {})
5297         try:
5298             if block_migration:
5299                 block_device_info = self._get_instance_block_device_info(
5300                     context, instance)
5301                 disk = self.driver.get_instance_disk_info(
5302                     instance, block_device_info=block_device_info)
5303             else:
5304                 disk = None
5305 
5306             pre_migration_data = self.compute_rpcapi.pre_live_migration(
5307                 context, instance,
5308                 block_migration, disk, dest, migrate_data)
5309             migrate_data['pre_live_migration_result'] = pre_migration_data

nova/virt/libvirt/driver.py

6418     def get_instance_disk_info(self, instance,
6419                                block_device_info=None):
6420         try:
6421             dom = self._host.get_domain(instance)
6422             xml = dom.XMLDesc(0)
6423         except libvirt.libvirtError as ex: 
6424             error_code = ex.get_error_code()
6425             msg = (_('Error from libvirt while getting description of '
6426                      '%(instance_name)s: [Error Code %(error_code)s] '
6427                      '%(ex)s') %
6428                    {'instance_name': instance.name,
6429                     'error_code': error_code,
6430                     'ex': ex})
6431             LOG.warn(msg)
6432             raise exception.InstanceNotFound(instance_id=instance.name)
6433 
6434         return jsonutils.dumps(
6435                 self._get_instance_disk_info(instance.name, xml,
6436                                              block_device_info))

This sets disk_info to an encoded JSON string (for example u'[{"foo":"bar},{"bar":"foo"}]'), causing the failure documented in c#0.

The get_instance_disk_info method switched back to plain strings for Liberty with the following change :

libvirt: Remove unnecessary JSON conversions
https://review.openstack.org/#/c/177437/6

Comment 2 Prasanth Anbalagan 2016-04-19 19:51:43 UTC
Verified as follows,

***********
VERSION
***********

[root@cougar14 ~(keystone_admin)]# yum list installed | grep openstack-nova
openstack-nova-api.noarch        2014.1.5-31.el7ost      @rhelosp-5.0-el7-puddle
openstack-nova-cert.noarch       2014.1.5-31.el7ost      @rhelosp-5.0-el7-puddle
openstack-nova-common.noarch     2014.1.5-31.el7ost      @rhelosp-5.0-el7-puddle
openstack-nova-conductor.noarch  2014.1.5-31.el7ost      @rhelosp-5.0-el7-puddle
openstack-nova-console.noarch    2014.1.5-31.el7ost      @rhelosp-5.0-el7-puddle
openstack-nova-novncproxy.noarch 2014.1.5-31.el7ost      @rhelosp-5.0-el7-puddle
openstack-nova-scheduler.noarch  2014.1.5-31.el7ost      @rhelosp-5.0-el7-puddle

***********
LOGS
***********

[root@cougar14 ~(keystone_admin)]# nova list
+--------------------------------------+------+--------+------------+-------------+---------------------+
| ID                                   | Name | Status | Task State | Power State | Networks            |
+--------------------------------------+------+--------+------------+-------------+---------------------+
| ee4768e6-f5b0-4d47-9592-42c4c07915e2 | vm1  | ACTIVE | -          | Running     | public=172.24.4.227 |
+--------------------------------------+------+--------+------------+-------------+---------------------+
[root@cougar14 ~(keystone_admin)]# nova show vm1 | grep host
| OS-EXT-SRV-ATTR:host                 | rhos-compute-node-02.lab.eng.rdu2.redhat.com             |
| OS-EXT-SRV-ATTR:hypervisor_hostname  | rhos-compute-node-02.lab.eng.rdu2.redhat.com             |
| hostId                               | 8ebb3b36bc169b837d1b05b0fc35d5ed6cac84d365e456474b110352 |
[root@cougar14 ~(keystone_admin)]# 
[root@cougar14 ~(keystone_admin)]# nova live-migration --block-migrate vm1 lynx13.qa.lab.tlv.redhat.com
[root@cougar14 ~(keystone_admin)]# nova list
+--------------------------------------+------+--------+------------+-------------+---------------------+
| ID                                   | Name | Status | Task State | Power State | Networks            |
+--------------------------------------+------+--------+------------+-------------+---------------------+
| ee4768e6-f5b0-4d47-9592-42c4c07915e2 | vm1  | ACTIVE | -          | Running     | public=172.24.4.227 |
+--------------------------------------+------+--------+------------+-------------+---------------------+
[root@cougar14 ~(keystone_admin)]# nova show vm1 | grep host
| OS-EXT-SRV-ATTR:host                 | lynx13.qa.lab.tlv.redhat.com                             |
| OS-EXT-SRV-ATTR:hypervisor_hostname  | lynx13.qa.lab.tlv.redhat.com                             |
| hostId                               | dfa3e1d7c98365b3561b2d5525eb5c6bdca87997df642cbbc06ce55a |

Comment 4 errata-xmlrpc 2016-04-26 15:39:41 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://rhn.redhat.com/errata/RHBA-2016-0690.html


Note You need to log in before you can comment on or make changes to this bug.