Bug 1958398
Summary: | "This VM has no graphic display device" error after upgrade from RHV 4.3 to RHV 4.4 for some VMs | ||
---|---|---|---|
Product: | Red Hat Enterprise Virtualization Manager | Reporter: | Robert McSwain <rmcswain> |
Component: | ovirt-engine | Assignee: | Shmuel Melamud <smelamud> |
Status: | CLOSED ERRATA | QA Contact: | Guilherme Santos <gdeolive> |
Severity: | high | Docs Contact: | |
Priority: | unspecified | ||
Version: | 4.4.4 | CC: | ahadas, akaiser, dfodor, gdeolive, mavital, michal.skrivanek |
Target Milestone: | ovirt-4.4.8 | Keywords: | ZStream |
Target Release: | --- | ||
Hardware: | x86_64 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | ovirt-engine-4.4.8.3 | Doc Type: | No Doc Update |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2021-09-08 14:12:04 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | Virt | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
Robert McSwain
2021-05-07 20:08:34 UTC
Looking at the database, the first problematic VM is set with: arik=# select type, device, is_managed, is_plugged from vm_device where vm_id='a79973de-278f-4252-a657-3d231291aef5'; type | device | is_managed | is_plugged ------------+---------------+------------+------------ video | qxl | t | f graphics | spice | t | t graphics | vnc | t | t controller | usb | t | t The fact that the video/qxl device is unplugged, together with the fix for bz 1729424 explains why the VM starts with: <video> <model type="none"/> </video> However, it is not clear why the video device of the imported VM is set as unplugged (it is set as unplugged during import for sure, since we see video with model.type=none at the first attempt to run the VM after it's imported). Need to think if this might be related somehow to bz 1844274 (In reply to Arik from comment #2) > (it is set as unplugged during import for sure, since we see video > with model.type=none at the first attempt to run the VM after it's imported). OK, so this part is incorrect. The device could not get unplugged during import, we import the devices as they are in the OVF. This means the video device must have been unplugged during the time the VM was in the 4.3 environment (one way that leads to that is running the VM with run-once + headless mode). In 4.3, in case a VM without no plugged device is started, the VM gets an unmanaged cirrus device (that's why from the user's point of view it appeared as it worked on the 4.3 environment). However, due to bz 1729424, such VM starts in 4.4 with video device with model=none and thus the console doesn't work. As video devices are not hot-pluggable, we can safely set them as plugged on import to overcome this issue. Verified on: ovirt-engine-4.4.8.1-0.9.el8ev.noarch Steps: 1. created two VMs on a 4.3.11 env 2. set one vm as headless mode and left the other vm as default console configuration 3. exported the vms and imported in the 4.4.8 env 4. run both vms Results: both vms started successfully (In reply to Guilherme Santos from comment #10) > Verified on: > ovirt-engine-4.4.8.1-0.9.el8ev.noarch > > Steps: > 1. created two VMs on a 4.3.11 env > 2. set one vm as headless mode and left the other vm as default console > configuration I'm afraid that doesn't produce the desired state - we want to have a video device set as unplugged in 4.3.11 when you set the VM as headless, there is no video device and with default console configuration there is a plugged video device. So can please check in the vm devices tab whether the video device of any of these VMs was unplugged? If it's not unplugged then a possible way to achieve this is by creating a VM with default console and then start it in run-once mode as headless > 3. exported the vms and imported in the 4.4.8 env > 4. run both vms > > Results: > both vms started successfully We don't only need to check that the VM starts but also that we can open a console to that VM Moving back to ON_QA for visibility Hi Arik, thank you for all the info. Indeed, I checked the video device in the 4.3.11 vm and it was plugged. I retested from scratch following the step you mentioned (VM with default configurations and run once as headless) and I confirmed that device was unplugged before I exported the VM. Then, I imported the vm in the 4.4.8 env successfully, also ran it and consoled on it successfully So verified (same versions as comment #10). 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 (RHV Manager (ovirt-engine) [ovirt-4.4.8]), 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://access.redhat.com/errata/RHBA-2021:3460 Due to QE capacity, we are not going to cover this issue in our automation |