Hi Gianluca, As mentioned in below RFE, EL 8.4 is not enough, you also need Advanced Virtualization 8.4 (libvirt >= 7.0.0) and AFAIK this has not been released upstream: https://bugzilla.redhat.com/show_bug.cgi?id=1933974 So right now you can use 4.6 cluster level only on hosts running on CentOS Stream. Also the error message tells you that all your Up hosts need to provide EL + AV 8.4 features before you upgrade cluster level. But of course you can upgrade them one by one, it's not required to put all hosts into Maintenance before upgrade. Regards, Martin On Tue, Jun 8, 2021 at 3:40 PM Gianluca Cecchi <gianluca.cecchi@gmail.com> wrote:

Hello, I updated all my 3 CentOS 4.4.5 hosts to 4.4.6. Or at least upgrading them from the GUI I see now they are on CentOS 8.4 and: [root@ov200 ~]# rpm -qa| grep 4.4.6 ovirt-release44-4.4.6.3-1.el8.noarch ovirt-host-4.4.6-1.el8.x86_64 ovirt-host-dependencies-4.4.6-1.el8.x86_64 [root@ov200 ~]#

How to see from the webadmin GUI or from the Host Console that they are indeed 4.4.6?

If I try to set cluster compatibility version to 4.6 I get:

Error while executing action: Cannot change Cluster Compatibility Version to higher version when there are active Hosts with lower version. -Please move Host ov300, ov301, ov200 with lower version to maintenance first.

I don't remember the need of giving downtime to get new cluster version...

Or what are further requirements (and benefits....) to upgrade to 4.6?

Thanks, Gianluca

_______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/privacy-policy.html oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/MKF7J6M2DC2H5Y...

-- Martin Perina Manager, Software Engineering Red Hat Czech s.r.o.

On Tue, Jun 8, 2021 at 4:10 PM Gianluca Cecchi <gianluca.cecchi@gmail.com> wrote:

On Tue, Jun 8, 2021 at 3:53 PM Martin Perina <mperina@redhat.com> wrote:

...

Hi Gianluca,

As mentioned in below RFE, EL 8.4 is not enough, you also need Advanced Virtualization 8.4 (libvirt >= 7.0.0) and AFAIK this has not been released upstream:

https://bugzilla.redhat.com/show_bug.cgi?id=1933974

So right now you can use 4.6 cluster level only on hosts running on CentOS Stream.

Also the error message tells you that all your Up hosts need to provide EL + AV 8.4 features before you upgrade cluster level. But of course you can upgrade them one by one, it's not required to put all hosts into Maintenance before upgrade.

Regards, Martin

OK, thanks for the note. Inside Doc Text of the bugzilla above there is: " New features available in compatibility 4.6 are tracked as separate bugs depending on this bug. " Have you the id of the bugs related to these new features provided in 4.4.6?

Relevant bugs are linked in Blocks field in Bugzilla: *1669178* <https://bugzilla.redhat.com/show_bug.cgi?id=1669178> - [RFE] Q35 SecureBoot - Add ability to preserve variable store certificates *1688177* <https://bugzilla.redhat.com/show_bug.cgi?id=1688177> - [RFE] [SR-IOV] [Tech Preview RHV 4.4.6] Migration should not require downtime as of today *1821199* <https://bugzilla.redhat.com/show_bug.cgi?id=1821199> - HP VM fails to migrate between identical hosts (the same cpu flags) not supporting TSC *1834250* <https://bugzilla.redhat.com/show_bug.cgi?id=1834250> - CPU hotplug on UEFI VM causes VM reboot *1892800* <https://bugzilla.redhat.com/show_bug.cgi?id=1892800> - NVDIMM: VM hangs in wait for launch for more than 7 minutes before VM is up *1906074* <https://bugzilla.redhat.com/show_bug.cgi?id=1906074> - [RFE] Support disks copy between regular and managed block storage domains *1927718* <https://bugzilla.redhat.com/show_bug.cgi?id=1927718> - [RFE] Provide Reset option for VMs *1936163* <https://bugzilla.redhat.com/show_bug.cgi?id=1936163> - Enable bochs-display for UEFI guests by default *1936164* <https://bugzilla.redhat.com/show_bug.cgi?id=1936164> - Enable KVM Software TPM by default *1950752* <https://bugzilla.redhat.com/show_bug.cgi?id=1950752> - [RFE][CBT] redefine only the checkpoint that the backup is taken from and not the entire chain

BTW [OT]: is RHV-H 4.4.6 host image ready for cluster level 4.6 or not yet?

Yes, oVirt Node 4.4.6 is based CentOS Stream: *1907833* <https://bugzilla.redhat.com/show_bug.cgi?id=1907833> - Rebase oVirt Node on CentOS Stream 8

Gianluca

-- Martin Perina Manager, Software Engineering Red Hat Czech s.r.o.

On Tue, Jun 8, 2021 at 6:14 PM Gianluca Cecchi <gianluca.cecchi@gmail.com> wrote:

On Tue, Jun 8, 2021 at 5:07 PM Martin Perina <mperina@redhat.com> wrote:

...

...

" Have you the id of the bugs related to these new features provided in 4.4.6?

Relevant bugs are linked in Blocks field in Bugzilla:

1669178 - [RFE] Q35 SecureBoot - Add ability to preserve variable store certificates 1688177 - [RFE] [SR-IOV] [Tech Preview RHV 4.4.6] Migration should not require downtime as of today 1821199 - HP VM fails to migrate between identical hosts (the same cpu flags) not supporting TSC 1834250 - CPU hotplug on UEFI VM causes VM reboot 1892800 - NVDIMM: VM hangs in wait for launch for more than 7 minutes before VM is up 1906074 - [RFE] Support disks copy between regular and managed block storage domains 1927718 - [RFE] Provide Reset option for VMs 1936163 - Enable bochs-display for UEFI guests by default 1936164 - Enable KVM Software TPM by default 1950752 - [RFE][CBT] redefine only the checkpoint that the backup is taken from and not the entire chain

OK

...

...

BTW [OT]: is RHV-H 4.4.6 host image ready for cluster level 4.6 or not yet?

Yes, oVirt Node 4.4.6 is based CentOS Stream:

1907833 - Rebase oVirt Node on CentOS Stream 8

I was referring to the downstream product with RHV-H 4.4.6... it is because of this that I used the [OT] I'm not sure if this 4.6 level is ready only upstream right now and needs further improvements to go downstream to RHV, or if it is already there production ready.

This is indeed off-topic, but I do not think we have to hide anything - everyone knows that almost all of oVirt's development is done by Red Hat people. For many years, the oVirt project struggled with our scheduling and release cycles - we wanted to be able to use and rely on newer features of RHEL, when available, but also to do a decent job for oVirt users, who do not have access to RHEL the moment it's GA. As developers, and also as Red Hatters, we already had access to internal development versions of RHEL, so could QA on them prior to releases (of both RHEL and oVirt/RHV). But what to do with oVirt users? It was always hard. We often ended up rebuilding RHEL SRPMs right after their release, in copr, before CentOS included them, for oVirt - including adding these to ovirt-release* rpms, and including having to later remove them from there once CentOS did include them - just so that oVirt users can use them before they are available in CentOS Linux. With the move to CentOS Stream, all of this is gone. oVirt is only building upon Stream, always has the latest bits, and everyone is happy! Well, at least in theory. In practice, everything is still new to all of us, but I think that generally speaking, things are working well. To answer your direct question: RHV 4.4.6 is based on RHEL+AV 8.4, and the release dates of each part were coordinated internally for this. Best regards, -- Didi

On Mon, Jul 26, 2021 at 7:38 PM Gianluca Cecchi <gianluca.cecchi@gmail.com> wrote:

On Wed, Jun 9, 2021 at 7:25 AM Yedidyah Bar David <didi@redhat.com> wrote:

...

To answer your direct question: RHV 4.4.6 is based on RHEL+AV 8.4, and the release dates of each part were coordinated internally for this.

Best regards, -- Didi

Just to confirm that today I updated one of my DCs from oVirt 4.4.6 to the latest async of 4.4.7 and I was able to set Cluster and DC levels to 4.6.

Thanks for the update! Best regards, -- Didi