- Users - oVirt List Archives

Is oVirt a dead project ???
by Diggy Mc 17 Sep '25

17 Sep '25

Based on the lack of list activity and updates, I am starting to think oVirt is a dead project. I sure hope not. It is a great product. My two cents: Whomever at Red Hat decided to abandon RHEV is an idiot. j/s Does anyone know if Oracle plans to fund the oVirt project? I think it would be a great move on their part. I believe virtualization still has a future for years to come. If oVirt is a dead project, what is the next best choice enterprise virtualization? ProxMox? Microsoft Hypervisor? Something else? Feedback is appreciated because I manage/recommend virtualization solutions for several small- and medium-sized companies.

5 4

please ignore, it's just a test
by Marc Dequènes (Duck) 17 Sep '25

17 Sep '25

a very taisty test -- Marc Dequènes (Duck) RedHat OSPO CommInfra

1 0

Is the ovirt permanently dead or will there be a new version?
by ecsi＠ecsi.hu 16 Sep '25

16 Sep '25

Hi all! There hasn't been a new release for almost two years... Someone should say something, but not empty promises.

3 2

Is the ovirt permanently dead or will there be a new version?
by ecsi＠ecsi.hu 15 Sep '25

15 Sep '25

Hi all! There hasn't been a new release for almost two years... Someone should say something, but not empty promises.

1 0

Is the ovirt permanently dead or will there be a new version?
by ecsi＠ecsi.hu 15 Sep '25

15 Sep '25

Hi all! There hasn't been a new release for almost two years... Someone should say something, but not empty promises.

1 0

Unable to launch new VM on few hosts due "Max free Memory for scheduling new VMs:" has less value
by dushyantk.sun＠gmail.com 11 Sep '25

11 Sep '25

Few hosts shows unable to launch vms due memory is not available to schedule vm. However vm has enough memory when check free command. If any one come across such issue, please let me know how we can fix it in oracle olvm

1 2

how to install perticular version of hosted engine such as Version 4.5.4-1.0.31.el8
by dushyantk.sun＠gmail.com 11 Sep '25

11 Sep '25

Currently I am testing upgrade of ovirt engine and looking for help on installing ovirt engine Version 4.5.4-1.0.31.el8. I tried to install this particular version however automated installation hosted engine deploy update it to 4.5.5., It fetches latest rpm from repo. Is there way we can install particular version of hosted engine instead of latest version. Required version of hosted engine : Version 4.5.4-1.0.31.el8

1 0

New oVirt plugin for The Foreman
by Frank Wall 01 Sep '25

01 Sep '25

Hi, for those of you using Foreman with oVirt: Please note that starting with Foreman 3.16 you have to install a new oVirt plugin. The plugin will be available from both the official Foreman DNF/DEB repositories, as well as from rubygems.org. For those of you that would like to support the new oVirt plugin, use the GitHub issue tracker to create issues and submit pull-requests: https://github.com/markt-de/foreman_ovirt/ Ciao - Frank

2 1

Re: ovirt 4.5 VNC Failed to complete handshake Error in the pull function on Windows
by Mark Atkinson 19 Aug '25

19 Aug '25

Were you able to get this working? To fix the problem: 1. Download the Engine CA Certificate (pki-resource) 2. Copy the Engine CA Certificate to /etc/pki/ca-trust/source/anchors 3. run #update-ca-trust extract 4. Restart your web browser and login to oVirt Manager 5. Check VM - Console settings 6. Launch Console - Open .vv file and you should be connected If it errors check DNS make sure you can resolve to your ovirt-engine. -Mark

1 0

oVirt 4.5.4-1.el8 - unable to add users
by Andrei Verovski 15 Aug '25

15 Aug '25

Hi, I have oVirt 4.5.4-1.el8, deleted expired user, and now unable to add any user / group. Typing user name, press “Go”, and the nothing happens. Restarted oVirt engine VM, no changes. How to fix this problem? Thanks in advance for any help. Andrei

1 0

PsqlException at Engine
by Christoph Koehler 08 Aug '25

08 Aug '25

Dear all, Since the last host update to the latest master snapshot, we have encountered a situation where hosts with the following CPU can no longer be activated: Xeon Gold 5418Y. The error comes from a database function of the engine: "PSQLEXCEPTION: ERROR: value too long for type character varying(4000) Where: SQL statement "UPDATE vds_dynamic SET cpuflags = vcpu_flags WHERE vdsid = vvds_id" PL/pgSQL function updatecpuflags(uuid, character varying) line 3 at SQL statement" Since the update, the length of the CPU flags exceeds the datatype varchar(4000) as declared in the DB. Currently, we have no solution, and the nodes remain 'unassigned'. Does anyone know of a workaround or when this presumably known error will be fixed? Thank you and best regards, Christoph

3 4

Fwd: [External] : LINSTOR Managed Block Storage Plugin Issue
by Simon Coter 06 Aug '25

06 Aug '25

Hi Jean, please, look at the GitHub issue below opened from LINBIT. Do you have any comment and/or suggestion on how this should be fixed ? Adding Shubha from our engineering. Simon Begin forwarded message: From: Ted Davis <ted.davis(a)oracle.com> Subject: FW: [External] : LINSTOR Managed Block Storage Plugin Issue Date: July 31, 2025 at 11:03:47 PM GMT+2 To: John Priest <john.priest(a)oracle.com>, Marcos Sungaila <marcos.sungaila(a)oracle.com> Cc: Simon Coter <simon.coter(a)oracle.com> John or Marcos, Does the below email from Linbit and the issue they raise of interest to OLVM team? It looks like it pertains to cinder plugin which I don’t think we support. Ted Davis | Sr. Oracle Linux & Oracle Virtualization ISV / IHV Manager | 801-601-9094 Oracle Linux & Oracle Virtualization Alliance Team Oracle, Inc. | Oracle Linux<https://www.oracle.com/linux/> | Oracle Virtualization<https://www.oracle.com/virtualization/> From: Ryan Ronnander <rronnander(a)linbit.com<mailto:rronnander@linbit.com>> Sent: Wednesday, July 2, 2025 10:10 AM To: Ted Davis <ted.davis(a)oracle.com<mailto:ted.davis@oracle.com>>; Marcos Sungaila <marcos.sungaila(a)oracle.com<mailto:marcos.sungaila@oracle.com>>; John Priest <john.priest(a)oracle.com<mailto:john.priest@oracle.com>> Cc: Moritz Wanzenböck <moritz.wanzenboeck(a)linbit.com<mailto:moritz.wanzenboeck@linbit.com>> Subject: [External] : LINSTOR Managed Block Storage Plugin Issue Hi everyone, I'm forwarding this issue your way in case there are any OLVM developer contacts you can relay this to if needed. Moritz, one of our developers who works on many of our platform integration plugins for LINSTOR, recently opened a GitHub issue in oVirt VDSM<https://urldefense.com/v3/__https:/github.com/oVirt/vdsm/issues/433__;!!ACW…> regarding storage volume behavior with our Cinderlib plugin. Long story short, after some testing we found that OLVM + LINSTOR has trouble migrating VMs back to virtual hosts that have failed (such as a forced "poweroff" with VMs running) after the hosts are up and operational again. It's been great working closely with all of you and everyone else from Oracle this year, and of course we want to make sure the LINSTOR storage plugin handles every expected use case in OLVM. Have a great weekend heading into the 4th of July. Thanks, Ryan

1 0

Re: [External] : Custom OMVF for VM
by Marcos Sungaila 11 Jul '25

11 Jul '25

Hi Domien, Possible it is, but you will need to test this deployment. The firmware json files are stored under /usr/share/qemu/firmware on each KVM host, and the fd files are stored under /usr/share/edk2/ovmf/ on the KVM hosts, too. As far as I understand, and you need to confirm it, the Chipset/Firmware Type option in the web UI is retrieved from the PostgreSQL database requiring you to add additional entries to it so the web portal will display the corresponding options to deploy this appliance. The good point, by going this path you will be able to create a great blog article describing how to create a custom entry in oVirt to deploy Sonicwall in it. Marcos From: Domien Van Rompaey via Users <users(a)ovirt.org> Sent: Tuesday, July 8, 2025 10:45 AM To: users(a)ovirt.org Subject: [External] : [ovirt-users] Custom OMVF for VM Hi We want to use the Sonicwall NSv appliance on our oVirt Cluster. Since version 7.1.1 they require the installation of a custom OMVF (UEFI Firmware). Sonicwall KB: https://www.sonicwall.com/support/knowledge-base/how-to-deploy-nsv-kvm-imag…<https://urldefense.com/v3/__https:/www.sonicwall.com/support/knowledge-base…> Is this possible on oVirt? I probably already know the answer (no) but I want to be sure. Met vriendelijke groeten, Domien Van Rompaey Security Engineer Cloud & ICT‑infrastructure | Security Tel +32 51 76 00 00 [https://website-api.dynamate.be/public/E-mailhandtekening/DYNAMATE-handteke…]<https://urldefense.com/v3/__https:/eu.content.exclaimer.net/?url=https*3A*2…> www.dynamate.be<https://urldefense.com/v3/__https:/eu.content.exclaimer.net/?url=https*3A*2…> [https://website-api.dynamate.be/public/E-mailhandtekening/instagram-60-150.…]<https://urldefense.com/v3/__https:/eu.content.exclaimer.net/?url=https*3A*2…> [https://website-api.dynamate.be/public/E-mailhandtekening/facebook-60-150.p…]<https://urldefense.com/v3/__https:/eu.content.exclaimer.net/?url=https*3A*2…> [https://website-api.dynamate.be/public/E-mailhandtekening/linkedin-60-150-2…]<https://urldefense.com/v3/__https:/eu.content.exclaimer.net/?url=https*3A*2…>

1 0

Custom OMVF for VM
by Domien Van Rompaey 08 Jul '25

08 Jul '25

Hi We want to use the Sonicwall NSv appliance on our oVirt Cluster. Since version 7.1.1 they require the installation of a custom OMVF (UEFI Firmware). Sonicwall KB: https://www.sonicwall.com/support/knowledge-base/how-to-deploy-nsv-kvm-imag… Is this possible on oVirt? I probably already know the answer (no) but I want to be sure. Met vriendelijke groeten, Domien Van Rompaey Security Engineer Cloud & ICT‑infrastructure | Security Tel +32 51 76 00 00 [https://website-api.dynamate.be/public/E-mailhandtekening/DYNAMATE-handteke…]<https://eu.content.exclaimer.net/?url=https%3A%2F%2Fwww.christiaens.net%2F&…> www.dynamate.be<https://eu.content.exclaimer.net?url=https%3A%2F%2Fwww.dynamate.be%2F&tenan…> [https://website-api.dynamate.be/public/E-mailhandtekening/instagram-60-150.…]<https://eu.content.exclaimer.net/?url=https%3A%2F%2Fwww.instagram.com%2Fwea…> [https://website-api.dynamate.be/public/E-mailhandtekening/facebook-60-150.p…]<https://eu.content.exclaimer.net/?url=https%3A%2F%2Fwww.facebook.com%2Fwear…> [https://website-api.dynamate.be/public/E-mailhandtekening/linkedin-60-150-2…]<https://eu.content.exclaimer.net/?url=https%3A%2F%2Fwww.linkedin.com%2Fcomp…>

1 0

Unable to start engine, vm status "Paused"
by mrtnz.om＠gmail.com 01 Jul '25

01 Jul '25

Hello everyone, the engine switched to paused state after attempting to place a host into maintenance, specifically when i selected the option to stop the gluster service. Glusterfs is 3 bricks [root@ovirt10 glusterfs]# gluster v heal engine info Brick 10.64.7.10:/gluster_bricks/engine/engine Status: Connected Number of entries: 0 Brick 10.64.7.11:/gluster_bricks/engine/engine Status: Connected Number of entries: 0 Brick 10.64.7.12:/gluster_bricks/engine/engine Status: Connected Number of entries: 0 --== Host ovirt11 (id: 3) status ==-- Host ID : 3 Host timestamp : 34188104 Score : 3400 Engine status : {"vm": "up", "health": "bad", "detail": "Paused", "reason": "bad vm status"} Cluster is in global maintenance The ovirt node i attempted to place in maintenance had no vms running (ovirt21) attempting to restart the engine when gluster is not running on ovirt21 results in the following error: MainThread::WARNING::2025-07-01 13:05:25,988::storage_broker::100::ovirt_hosted_engine_ha.broker.storage_broker.StorageBroker::(__init__) Can't connect vdsm storage: path to storage domain 4d31ba9e-5167-4af8-bc2a-361cce9b8278 not found in /rhev/data-center/mnt/glusterSD If i restart glusterd, and ovirt-agent and ovirt-broker on the host i put into maintenance (ovirt21), then attempt to restart the engine vm, it starts up but then is paused shortly after when glusterd service is automatically stopped on ovirt node 21. It seems the engine is stuck trying to put that host into maintenance, as soon as the engine vm starts it's paused when glusterd is stopped on ovirt21. Is there a way to safely unmount the engine gluster volume from the empty host without affecting the rest of the cluster? and/or, is there a way to stop the engine from putting that host into maintenance? Thanks! Mario M

1 0

Unable to start engine, vm status "Paused"
by mrtnz.om＠gmail.com 01 Jul '25

01 Jul '25

Hello everyone, the engine switched to paused state after attempting to place a host into maintenance, specifically when i selected the option to stop the gluster service. Glusterfs is 3 bricks [root@ovirt10 glusterfs]# gluster v heal engine info Brick 10.64.7.10:/gluster_bricks/engine/engine Status: Connected Number of entries: 0 Brick 10.64.7.11:/gluster_bricks/engine/engine Status: Connected Number of entries: 0 Brick 10.64.7.12:/gluster_bricks/engine/engine Status: Connected Number of entries: 0 --== Host ovirt11 (id: 3) status ==-- Host ID : 3 Host timestamp : 34188104 Score : 3400 Engine status : {"vm": "up", "health": "bad", "detail": "Paused", "reason": "bad vm status"} Cluster is in global maintenance The ovirt node i attempted to place in maintenance had no vms running (ovirt21) attempting to restart the engine when gluster is not running on ovirt21 results in the following error: MainThread::WARNING::2025-07-01 13:05:25,988::storage_broker::100::ovirt_hosted_engine_ha.broker.storage_broker.StorageBroker::(__init__) Can't connect vdsm storage: path to storage domain 4d31ba9e-5167-4af8-bc2a-361cce9b8278 not found in /rhev/data-center/mnt/glusterSD If i restart glusterd, and ovirt-agent and ovirt-broker on the host i put into maintenance (ovirt21), then attempt to restart the engine vm, it starts up but then is paused shortly after when glusterd service is automatically stopped on ovirt node 21. It seems the engine is stuck trying to put that host into maintenance, as soon as the engine vm starts it's paused when glusterd is stopped on ovirt21. Is there a way to safely unmount the engine gluster volume from the empty host without affecting the rest of the cluster? and/or, is there a way to stop the engine from putting that host into maintenance? Thanks! Mario M

1 0

Ovirt 4.5.5 + Veeam Unable to get path commander
by victordanco＠hotmail.com 01 Jul '25

01 Jul '25

Good morning dear community. We are experiencing a problem when taking a backup through Veeam backup. I share the logs: Engine.log 2025-06-03 16:09:52,158-05 ERROR [org.ovirt.engine.core.sso.service.SsoService] (default task-2019) [53762455] OAuthException invalid_grant: The provided authorization grant for the auth code has expired. 2025-06-03 16:09:52,160-05 ERROR [org.ovirt.engine.core.aaa.filters.SsoRestApiAuthFilter] (default task-2006) [] Cannot authenticate using authentication Headers: invalid_grant: The provided authorization grant for the auth code has expired. 2025-06-03 16:09:52,429-05 INFO [org.ovirt.engine.core.sso.service.AuthenticationService] (default task-2006) [] User admin@internal-authz with profile [internal] successfully logged in with scopes: ovirt-app-admin ovirt-app-api ovirt-app-portal ovirt-ext=auth:sequence-priority=~ ovirt-ext=revoke:revoke-all ovirt-ext=token-info:authz-search ovirt-ext=token-info:public-authz-search ovirt-ext=token-info:validate ovirt-ext=token:password-access 2025-06-03 16:09:52,518-05 INFO [org.ovirt.engine.core.bll.aaa.CreateUserSessionCommand] (default task-2006) [60f224da] Running command: CreateUserSessionCommand internal: false. 2025-06-03 16:09:52,530-05 INFO [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (default task-2006) [60f224da] EVENT_ID: USER_VDC_LOGIN(30), User admin@internal-authz connecting from '192.168.170.52' using session 'dYQlOCMThsTOInKjX8GKWUpAk/SFspOh5vuX0efK1slxzWloBoK9jGqt7TwOapfxUR1kzGg0ctoofUC7/pjeVg==' logged in. 2025-06-03 16:09:52,784-05 WARN [org.ovirt.engine.api.restapi.util.LinkHelper] (default task-2006) [] Can't find relative path for class "org.ovirt.engine.api.resource.StorageDomainVmDiskAttachmentsResource", will return null 2025-06-03 16:09:52,784-05 WARN [org.ovirt.engine.api.restapi.util.LinkHelper] (default task-2006) [] Can't find relative path for class "org.ovirt.engine.api.resource.StorageDomainVmDiskAttachmentsResource", will return null 2025-06-03 16:10:05,258-05 INFO [org.ovirt.engine.core.sso.servlets.OAuthRevokeServlet] (default task-2006) [] User admin@internal-authz with profile [internal] successfully logged out 2025-06-03 16:10:05,267-05 INFO [org.ovirt.engine.core.bll.aaa.TerminateSessionsForTokenCommand] (default task-2019) [78785b84] Running command: TerminateSessionsForTokenCommand internal: true. 2025-06-03 16:10:05,305-05 INFO [org.ovirt.engine.core.sso.servlets.OAuthRevokeServlet] (default task-2006) [] User admin@internal-authz with profile [internal] successfully logged out 2025-06-03 16:10:05,310-05 INFO [org.ovirt.engine.core.bll.aaa.TerminateSessionsForTokenCommand] (default task-2019) [4a6cfc6] Running command: TerminateSessionsForTokenCommand internal: true. 2025-06-03 16:10:28,401-05 INFO [org.ovirt.engine.core.sso.service.AuthenticationService] (default task-2006) [] User admin@internal-authz with profile [internal] successfully logged in with scopes: ovirt-app-admin ovirt-app-api ovirt-app-portal ovirt-ext=auth:sequence-priority=~ ovirt-ext=revoke:revoke-all ovirt-ext=token-info:authz-search ovirt-ext=token-info:public-authz-search ovirt-ext=token-info:validate ovirt-ext=token:password-access 2025-06-03 16:10:28,453-05 INFO [org.ovirt.engine.core.bll.aaa.CreateUserSessionCommand] (default task-2006) [152dcb43] Running command: CreateUserSessionCommand internal: false. 2025-06-03 16:10:28,466-05 INFO [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (default task-2006) [152dcb43] EVENT_ID: USER_VDC_LOGIN(30), User admin@internal-authz connecting from '192.168.170.52' using session '1WDL5jCeJNVnQCOSAJu8hd+M1IJJp5AChx3SQnpNjG+v2cZHNRYBsrQpmfhNAuB5KAHUun7HKUxbRQcs5OCalQ==' logged in. 2025-06-03 16:10:50,031-05 WARN [org.ovirt.engine.core.utils.virtiowin.VirtioWinReader] (EE-ManagedScheduledExecutorService-engineScheduledThreadPool-Thread-19) [6b0bdae7] Directory '/usr/share/virtio-win' doesn't exist. --------- Veeam Proxy: 2025-06-03 16:10:21.0156 00023 [194115] ERROR | [RhvCluster][myengine.co]: <== Request "Get" "https://myengine.co/ovirt-engine/api/vms/0a5b4e0e-477c-494a-ba43-1e6c48565a…", body: "{}" 2025-06-03 16:10:21.0156 00023 [194115] ERROR | [RhvCluster][myengine.co]: ==> Response "Get" "https://myengine.co/ovirt-engine/api/vms/0a5b4e0e-477c-494a-ba43-1e6c48565a…", "code: 401", duration: "163 msec", body: "[binary data]" Note that the user I use is admin@internal. I have read other forums that it is recommended to use admin@ovirt@internalsso or others, but veeam does not allow me to use admin@ovirt@internalsso. I appreciate your help or recommendations. Regards!

2 2

oVirt 4.5.7 - Rocky 9.4 - Update Quandry
by Clint Boggio 24 Jun '25

24 Jun '25

Good Day All - We're looking to run some updates to our oVirt environment and we're getting an error regarding selinux-policy and it's available version: Error: Problem 1: cannot install the best update candidate for package ovirt-vmconsole-1.0.9-3.el9.noarch - nothing provides selinux-policy >= 38.1.56-1.el9 needed by ovirt-vmconsole-1.0.9-3.20250519152134.git4943ea2.el9.noarch from copr:copr.fedorainfracloud.org:ovirt:ovirt-master-snapshot - nothing provides selinux-policy-base >= 38.1.56-1.el9 needed by ovirt-vmconsole-1.0.9-3.20250519152134.git4943ea2.el9.noarch from copr:copr.fedorainfracloud.org:ovirt:ovirt-master-snapshot Problem 2: package ovirt-vmconsole-host-1.0.9-3.20250519152134.git4943ea2.el9.noarch from copr:copr.fedorainfracloud.org:ovirt:ovirt-master-snapshot requires ovirt-vmconsole = 1.0.9-3.20250519152134.git4943ea2.el9, but none of the providers can be installed - cannot install the best update candidate for package ovirt-vmconsole-host-1.0.9-3.el9.noarch - nothing provides selinux-policy >= 38.1.56-1.el9 needed by ovirt-vmconsole-1.0.9-3.20250519152134.git4943ea2.el9.noarch from copr:copr.fedorainfracloud.org:ovirt:ovirt-master-snapshot - nothing provides selinux-policy-base >= 38.1.56-1.el9 needed by ovirt-vmconsole-1.0.9-3.20250519152134.git4943ea2.el9.noarch from copr:copr.fedorainfracloud.org:ovirt:ovirt-master-snapshot Problem 3: package python3.11-ovirt-imageio-client-2.5.1-0.202504080926.gitfde3ef0.el9.x86_64 from copr:copr.fedorainfracloud.org:ovirt:ovirt-master-snapshot requires python3.11-ovirt-engine-sdk4, but none of the providers can be installed Any guidance would be appreciated

1 0

Re: Network Performance Issue on oVirt Node Using Intel Platinum 8562Y+ (Sapphire Rapids) CPU
by domien.vanrompaey＠dynamate.be 24 Jun '25

24 Jun '25

Hi My issue was solved by updating the firmware on our Broadcom BCM57504 NIC (buggy firmware was 23.11.16.22). I also had to add the following option via ethtool for the NICs that were serving the bond: --offload eno12419np2 gro off --offload eno12429np3 gro off . This completely solved my problems. Maybe this can be relevant for your issue too.

1 0

Veeam - Move data over other network than ovirtmgmt
by domien.vanrompaey＠dynamate.be 18 Jun '25

18 Jun '25

Hi all I was wondering if anyone managed to get Veeam working with a dedicated backup VLAN instead of sending traffic over the ovirtmgmt interface. We have an OLVM setup (version 4.5.5-1.42.el8) with 3 hosts in 1 cluster for now. Each host has a 1Gbps NIC pair in a seperate management LAN that is only used for ovirtmgmt Each host also has a 10Gbps NIC Bond where all the VLAN's are trunked including a dedicated Backup VLAN. The hosts do not have an IP in this Backup VLAN configured (is this needed?) Each host also has 2 NICs dedicated to storage backend (iSCSI Multipathing) Our Veeam Appliance and workers are setup in the Backup VLAN and they can communicate with the backup repository that is also in this VLAN. When we take a backup, however, we see that the ovirtmgmt network is maxed out on trafic, which probably shouldn't happen. We come from a VMWare environment where the backup proxies handle all the trafic through the backup VLAN without relying on the management network and I was expecting it to work similarly Has anyone managed to get this working as I envision it? Or is the solution to simply link ovirtmgmt to the 10Gbps bond and skip the dedicated mgmt networking? I've also opened a ticket with Veeam for this, but for now I'm just getting questions from them and no real answers. If I get a useful solution from them I will also post it here

4 3

Network Performance Issue on oVirt Node Using Intel Platinum 8562Y+ (Sapphire Rapids) CPU
by dhayanands＠botvfx.com 17 Jun '25

17 Jun '25

Hi team, We’re facing a strange issue with network performance on one of our oVirt hosts and would like to get insights or guidance on possible misconfigurations or compatibility problems. Environment Details: oVirt Version: 4.5.5 Cluster CPU Type: Intel IceLake Server Family (default) Host A (Issue observed): CPU: Intel Xeon Platinum 8562Y+ (Sapphire Rapids) CPU Family/Model: Family 6 / Model 207 / Stepping 2 Host B (Working fine): CPU: Intel Xeon Gold 6354 (Ice Lake) CPU Family/Model: Family 6 / Model 106 / Stepping 6 Problem Description: We have a VM (kitsu) that was initially hosted on botovirtnode-006 (Platinum 8562Y+). The VM showed very poor download speeds (~10 Mbps), while upload was fine (~680 Mbps). After migrating the same VM to another host (botovirtnode-001 with Gold 6354 CPU), network performance drastically improved: Test On Affected Node On Healthy Node Download ~10 Mbps ~920 Mbps Upload ~680 Mbps ~760 Mbps This clearly indicates a host-specific issue, most likely related to hardware compatibility or CPU feature masking. Troubleshooting Done: Verified VM configurations (same on both hosts). Tested with speedtest-cli multiple times to confirm. Confirmed that VM virtual NIC, vNIC profile, and vSwitch settings are identical. Both hosts are part of the same cluster using the default Intel IceLake Server CPU type. Suspect the issue is due to CPU feature mismatch or unsupported stepping/model in the cluster’s CPU type.

3 3

Re: Network Performance Issue on oVirt Node Using Intel Platinum 8562Y+ (Sapphire Rapids) CPU
by Domien Van Rompaey 16 Jun '25

16 Jun '25

Hi there I was just about to post a similar issue when I saw your post. We have a cluster of 3 AMD EPYC 9454 48-Core Processor hosts who have BCM57504 NICs in a LACP bond to 2 Dell S5248F-ON switches. We are seeing assymetrical network performance between VM's making the cluster unusable. In previous test setups we did not have this issue. The issues seems to manifest only for traffic between hosts over VLAN networks and SDNs do not seem to be affected. If we do an upload test over a virtual firewall we get the same numbers as you do. If we do some iperf tests we get this: (We will use the name VM1 and VM2 who are in the same VLAN network. For hosts we will use Host1 and Host2 in the names) VM1: 10.22.21.10 VM2: 10.22.21.20 VM1 (Host2) to VM2 (Host1) [cid:image002.png@01DBDEA1.6BFA2C60] VM2 (Host1) -> VM1 (Host2) [cid:image003.png@01DBDEA1.6BFA2C60] When we put these on the same host (Host2) VM2 -> VM1 [cid:image004.png@01DBDEA1.6BFA2C60] VM1->VM2 [cid:image005.png@01DBDEA1.6BFA2C60] I’m believing the underlying issue will be the same. If it’s not I will probably split my issue in a seperate post. We raised a ticket with Oracle Support for this (we are running OLVM, latest version) but it’s going very slow. The issue is stopping us from bringing our cluster into production. Met vriendelijke groeten, Domien Van Rompaey Security Engineer Cloud & ICT‑infrastructure | Security Tel +32 51 76 00 00 [https://website-api.dynamate.be/public/E-mailhandtekening/DYNAMATE-handteke…]<https://eu.content.exclaimer.net/?url=https%3A%2F%2Fwww.christiaens.net%2F&…> www.dynamate.be<https://eu.content.exclaimer.net?url=https%3A%2F%2Fwww.dynamate.be%2F&tenan…> [https://website-api.dynamate.be/public/E-mailhandtekening/instagram-60-150.…]<https://eu.content.exclaimer.net/?url=https%3A%2F%2Fwww.instagram.com%2Fwea…> [https://website-api.dynamate.be/public/E-mailhandtekening/facebook-60-150.p…]<https://eu.content.exclaimer.net/?url=https%3A%2F%2Fwww.facebook.com%2Fwear…> [https://website-api.dynamate.be/public/E-mailhandtekening/linkedin-60-150-2…]<https://eu.content.exclaimer.net/?url=https%3A%2F%2Fwww.linkedin.com%2Fcomp…>

1 0

Unable to Start the Hosted Engine and the WebUI
by kelwintai＠proton.me 13 Jun '25

13 Jun '25

Hi there, I have tried to restart the Hosted Engine but I cannot ssh/ping the Hosted Engine after the restart. The oVirt webUI was down and I cannot start the VM in both GUI and CLI. I have used command "hosted-engine --console" to start a console but error message "Cannot find character device <null>" was prompted. I also checked the status of the Hosted Engine and the its state was "EngineStarting". However, the Hosted Engine will keep restarting in the same host automatically. Please help 1. Is there any commands that I can use to start the VM inside the Hosted Engine CLI. 2. Does anyone know why my Hosted Engine cannot be started successfully and accessed via CLI. Thank you for your help. KT

2 2

Please let me know how to resolve the below error!
by d_hashima＠sagaidc.co.jp 12 Jun '25

12 Jun '25

Error Content [ ERROR ] fatal: [localhost -> 192.168.1.252]: FAILED! => {"changed": false, "msg": "Failed to download metadata for repo 'centos-ceph-pacific': Cannot prepare internal mirrorlist: Curl error (6): Couldn't resolve host name for http://mirrorlist.centos.org/?release=8-stream&arch=x86_64&repo=storage-cep… [Could not resolve host: mirrorlist.centos.org]", "rc": 1, "results": []} environment CentOS Stream release ovirt-node-ng-installer-4.5.5-2023113015.el9.iso I am building it with ovirt-node-ng-installer-4.5.5-2023113015.el9.iso and plan to install oVirt Engine there. I am building it on a virtual server.

5 10

What is the future of oVirt?
by Giacomo Failla 10 Jun '25

10 Jun '25

Hi everyone, I’m writing this post because I think many of us are asking the same questions these days. Over the past few months, we’ve seen some big changes in the open source virtualization world. Red Hat’s decision to discontinue RHV and slow down upstream oVirt development has understandably created a lot of uncertainty — especially for those of us running critical infrastructure on top of oVirt. In this changing landscape, I’m wondering: how is the oVirt project doing today? Is there still an active roadmap? Are there developers and organizations continuing to contribute? And most importantly — what can we as a community do to help ensure that oVirt remains a solid and well-maintained option in the years ahead? At the end of the day, many of us need to make decisions about how to evolve our IT environments, and having clarity on whether oVirt will continue to be a reliable choice is really important. I’d love to hear from anyone close to the project or involved in development. Any updates, insights, or thoughts would be greatly appreciated. Thanks to everyone for your work and contributions to this community. Best regards, Giacomo Failla Head of Technology — Cheope SA

2 1