Bug 1396108

Description of problem:

The Administration Portal allows a user to try to migrate the HE VM Disk. It fails on Validation, as expected, but the Disk remains in Locked State and now the VM cannot be Migrated anymore.

Version-Release number of selected component (if applicable):
ovirt-engine-4.0.4

How reproducible:
100%

Steps to Reproduce:
1. Click on Hosted-Engine VM
2. Click on Disks Tab
3. Select Disk
4. Click Move
5. Select Destination, Click OK
6. Fails (Expected and Correct)
7. Disk Remains in Locked State
8. Click Migrate -> Fails
8. VM cannot migrate, it's stuck:
Cannot migrate VM: The following disks are locked: virtio-disk0. Please try again in a few minutes.

Actual results:
It fails when Validating (as expected), but should not leave the Disk in Locked State

Expected results:
Fails or Blocked. Disk in OK state.

This comment was originaly posted by gveitmic

Please attach Engine logs.
Also, which validation is failing that is expected to fail?

This comment was originaly posted by tnisan

Hi Tal,

The image is left locked at the end of the process.

engine=# select imagestatus from images where image_group_id = '55e10183-51d7-4f74-8cef-8fe21cd4f5ed';
 imagestatus 
-------------
           2
(1 row)

* Validation: the validation step for executing the disk move (snapshot for live migration), in the engine.

2016-11-13 22:03:42,779 WARN  [org.ovirt.engine.core.bll.snapshots.CreateAllSnapshotsFromVmCommand] (org.ovirt.thread.pool-6-thread-41) [31ed43b8] Validation of action 'CreateAllSnapshotsFromVm' failed for user admin@internal-authz. Reasons: VAR__ACTION__CREATE,VAR__TYPE__SNAPSHOT,ACTION_TYPE_FAILED_CANNOT_RUN_ACTION_ON_NON_MANAGED_VM

For logs, look at the end of the file I'm attaching.

This comment was originaly posted by gveitmic

Verification failed:
The HostedEngine VM's disk wasn't migrated and remain in locked state

from engine.log:
2016-11-23 10:46:25,444 INFO  [org.ovirt.engine.core.bll.storage.disk.MoveDisksCommand] (default task-32) [d18c5a1] Running command: MoveDisksCommand internal: false. Entities affected :  ID: 7361b89b-d614-4294-9f8c-cfb6c8be8c74 Type: DiskAction group CONFIGURE_DISK_STORAGE with role type USER
2016-11-23 10:46:25,572 INFO  [org.ovirt.engine.core.bll.storage.lsm.LiveMigrateVmDisksCommand] (default task-32) [d18c5a1] Lock Acquired to object 'EngineLock:{exclusiveLocks='[7361b89b-d614-4294-9f8c-cfb6c8be8c74=<DISK, ACTION_TYPE_FAILED_DISK_IS_BEING_MIGRATED$DiskName virtio-disk0>]', sharedLocks='[2230569e-9357-4ab2-915a-767fa5532db1=<VM, ACTION_TYPE_FAILED_OBJECT_LOCKED>]'}'
2016-11-23 10:46:25,872 INFO  [org.ovirt.engine.core.bll.storage.lsm.LiveMigrateVmDisksCommand] (org.ovirt.thread.pool-6-thread-35) [d18c5a1] Running command: LiveMigrateVmDisksCommand internal: false. Entities affected :  ID: 7361b89b-d614-4294-9f8c-cfb6c8be8c74 Type: DiskAction group DISK_LIVE_STORAGE_MIGRATION with role type USER
2016-11-23 10:46:26,127 WARN  [org.ovirt.engine.core.bll.snapshots.CreateAllSnapshotsFromVmCommand] (org.ovirt.thread.pool-6-thread-35) [6f3710c] Validation of action 'CreateAllSnapshotsFromVm' failed for user admin@internal-authz. Reasons: VAR__ACTION__CREATE,VAR__TYPE__SNAPSHOT,ACTION_TYPE_FAILED_CANNOT_RUN_ACTION_ON_NON_MANAGED_VM
2016-11-23 10:46:26,415 INFO  [org.ovirt.engine.core.bll.SerialChildCommandsExecutionCallback] (DefaultQuartzScheduler6) [6f3710c] Command 'LiveMigrateVmDisks' id: 'a9dad97f-61da-40cd-a42f-587f1177ba3b' child commands '[31c37cc5-aada-471f-9b3e-c335ff4dc52a]' executions were completed, status 'FAILED'
2016-11-23 10:46:27,518 ERROR [org.ovirt.engine.core.bll.storage.lsm.LiveMigrateVmDisksCommand] (DefaultQuartzScheduler1) [6f3710c] Ending command 'org.ovirt.engine.core.bll.storage.lsm.LiveMigrateVmDisksCommand' with failure.
2016-11-23 10:46:27,519 INFO  [org.ovirt.engine.core.bll.storage.lsm.LiveMigrateVmDisksCommand] (DefaultQuartzScheduler1) [6f3710c] Lock freed to object 'EngineLock:{exclusiveLocks='[7361b89b-d614-4294-9f8c-cfb6c8be8c74=<DISK, ACTION_TYPE_FAILED_DISK_IS_BEING_MIGRATED$DiskName virtio-disk0>]', sharedLocks='[2230569e-9357-4ab2-915a-767fa5532db1=<VM, ACTION_TYPE_FAILED_OBJECT_LOCKED>]'}'

logs attached

Created attachment 1223327 [details]
engine log

Raz, which Engine build did you run it on?

4.0.6.1-0.1.el7ev

Every LSM operation fails with:
"Error while executing action: User is not logged in."

Verified on 4.0.6.3-0.1.el7ev

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-2017-0043.html