network out of sync
by Nathanaël Blanchet
Hi,
After reinstalling ovirt hosts with alma Linux 9, all my hosts are out
of sync on the management bridge. I tried to sync them, but it has no
effects. Then I tried "static" and "DHCP" but it has no effect as well.
I precise that I was using ovirt node 4.5.5 el9, and the bug wasn't
present.
It seems to be a bug because, network is okay and there is no other
issue using alamlinux 9.5 with latest vdsmd.
Thank you for helping.
--
Nathanaël Blanchet
Administrateur Systèmes et Réseaux
Service Informatique et REseau (SIRE)
Département des systèmes d'information
227 avenue Professeur-Jean-Louis-Viala
34193 MONTPELLIER CEDEX 5
Tél. 33 (0)4 67 54 84 55
Fax 33 (0)4 67 54 84 14
blanchet(a)abes.fr
30 minutes
Unable to set up host bond networks
by erik＠linux.com
Hi
We are trying to create a cluster using OLVM 4.5 (with an up to date
OEL8 kernel) with a bond of 2 virtual interfaces (mode 2 XOR) that we
have set up with NetworkManager prior to adding the KVM host to oVirt.
Hosts are added successfully and we can see the two physical NICs
(eth0/eth1) in the Setup Host Networks dialog, however the bond0 device
is hidden as are the virtual slave interfaces (so we cannot create the
bond in oVirt either).
It appears the bond0 is sent to oVirt by VDSM as we see it is
continually sent in the vdsm log file from the host as well as listed
in the capabilities using vdsm-client. We didn't find obvious errors in
ovirt-engine.log or in host-deploy or in the vdsm logs on the KVM hosts.
We assume there is some validation or filtering oVirt uses that is
hiding the bond / slave interfaces and that maybe our slave interfaces
are missing some property oVirt wants but we cannot find it in the code
in order to work around this - any pointers appreciated.
Thanks
Erik
1 day, 19 hours
How to modify the MAC address of a bonding interface after changing the nic card
by wodel youchi
Hi,
ovirt 4.4 on centos 8.
We have changed the network card of a hypervisor which uses bonding
interface.
The bonding interface still uses the old MAC address of the former slave
interface.
We tried to modify it via CLI but it didn't work, as soon as the node
(hypervisor) is activated, the old MAC is reestablished.
it seems to be backed up somewhere.
Should we reinstall the host?
Regards.
1 day, 23 hours
Re: Unable to get console of self hosted engine via oracle kvm
by Dushyant Khobragade
For me it is not going any further.
On Fri, Mar 14, 2025 at 5:54 PM Geoff O'Callaghan <geoffocallaghan(a)gmail.com>
wrote:
> I didn't have to perform any extra steps, but I was using oVirt Node
> (current), not OLVM itself - I don't have an OLVM environment to test
> against, but might be able to build one over the next few days.
>
> I merely ssh'd into the host where the self-hosted engine was running and
> ran:
>
> *hosted-engine --console*
>
> That's it
>
> ------------------------------
> *From:* dushyantk.sun(a)gmail.com <dushyantk.sun(a)gmail.com>
> *Sent:* Thursday, March 13, 2025 7:02 AM
> *To:* users(a)ovirt.org <users(a)ovirt.org>
> *Subject:* [ovirt-users] Re: Unable to get console of self hosted engine
> via oracle kvm
>
> Yes I did.. But it is like stuck and not giving any prompt.
>
> Did we need to set terminal on kvm side.
>
> Also do you have any steps to perform prior to taking console?
> _______________________________________________
> Users mailing list -- users(a)ovirt.org
> To unsubscribe send an email to users-leave(a)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/ILVEWHD3J4D...
>
3 days, 3 hours
Unable to get console of self hosted engine via oracle kvm
by dushyantk.sun＠gmail.com
Hi,
I am trying to get console of hosted engine via KVM host where this SHE is.
I am getting below but unable to see login prompt. I have set the password for console as per oracle doc.
usecase: SHE UI is down so wanted to know whats going on SHE level by login into it.
# echo $TERM
xterm-256color
# hosted-engine --console
The engine VM is running on this host
Escape character is ^]
Not getting any prompt after this.
Anyone come across such scenario.
1 week
Re: Couldn't resolve host name for http://mirrorlist.centos.org/?release=8-stream&arch=x86_64&repo=storage-ceph-pacific
by Geoff O'Callaghan
Correcting my email - you can remove the following on an EL9 or related system. I haven't checked to see what happens on RHEL9 itself, but will later
**remove**
cat >/etc/yum.repos.d/CentOS-Stream-Extras-common.repo <<'EOF'
[c9s-extras-common]
name=CentOS Stream $releasever - Extras packages
metalink=https://mirrors.centos.org/metalink?repo=centos-extras-sig-extra...
gpgkey=https://www.centos.org/keys/RPM-GPG-KEY-CentOS-SIG-Extras
gpgcheck=1
repo_gpgcheck=0
metadata_expire=6h
countme=1
enabled=1
EOF
echo "9-stream" > /etc/yum/vars/stream
dnf distro-sync --nobest
**remove**
Summary is
## Installing on RHEL 9.0 derivatives
Once a minimal RHEL 9.0 derivative system has been built, you can install the pre-production master snapshots (oVirt Release 4.5.6 + Engine 4.5.7)
```bash
dnf copr enable -y ovirt/ovirt-master-snapshot centos-stream-9
dnf install -y ovirt-release-master
```
At this point you can choose to install the :
* Hosted Engine, `dnf install -y ovirt-hosted-engine-setup`
* A minimal host, `dnf install -y ovirt-host`
NOTE: This is using the latest builds. There hasn't been a release for some time, so your best chance of success is to use these snapshots. When there is a release i'll update the document to include the details.
________________________________
From: Geoff O'Callaghan
Sent: Wednesday, March 05, 2025 5:15 PM
To: users(a)ovirt.org; fiorletta(a)ssolo.eu
Subject: Re: [ovirt-users] Re: Couldn't resolve host name for http://mirrorlist.centos.org/?release=8-stream&arch=x86_64&repo=storage-c...
Let me try and break this down into a sequence of steps and let's see where it breaks down for you. Let's start with the simplest scenario. Centos 9 Stream system (6 core, 32Gb memory, 250Gb disk, single 1GbE nic - note i'm doing this in a virt-enabled VM, but bare metal should be similar). This is the sort of information I need to debug your problem.
*
Build a vanilla Centos 9 Stream system. (non-oVirt node image, but we can do that if you like)
*
My centos 9 system is called ovnode01.<fqdn> 192.168.1.111 and the manager i'm going to install is emanager.<fqdn> 192.168.1.251, both fqdn's resolve in my DNS (important)
*
As per https://ovirt.org/download/install_on_rhel.html
cat >/etc/yum.repos.d/CentOS-Stream-Extras-common.repo <<'EOF'
[c9s-extras-common]
name=CentOS Stream $releasever - Extras packages
metalink=https://mirrors.centos.org/metalink?repo=centos-extras-sig-extra...
gpgkey=https://www.centos.org/keys/RPM-GPG-KEY-CentOS-SIG-Extras
gpgcheck=1
repo_gpgcheck=0
metadata_expire=6h
countme=1
enabled=1
EOF
echo "9-stream" > /etc/yum/vars/stream
dnf distro-sync --nobest
*
*
dnf copr enable -y ovirt/ovirt-master-snapshot centos-stream-9
dnf install -y ovirt-release-master
*
You should get something like this
Running transaction
Preparing : 1/1
Installing : tar-2:1.34-7.el9.x86_64 1/2
Installing : ovirt-release-master-4.5.6-0.0.master.20240719070214.gitf7ad2c3.el9.noarch 2/2
Running scriptlet: ovirt-release-master-4.5.6-0.0.master.20240719070214.gitf7ad2c3.el9.noarch 2/2
Verifying : ovirt-release-master-4.5.6-0.0.master.20240719070214.gitf7ad2c3.el9.noarch 1/2
Verifying : tar-2:1.34-7.el9.x86_64
*
dnf install ovirt-hosted-engine-setup
*
Apart from this message 'Repository copr:copr.fedorainfracloud.org:ovirt:ovirt-master-snapshot is listed more than once in the configuration' you should have no other repository issues at this time.
*
663 packages later..... (my initial builds are minimal installs)
*
*
Remembering the above comment about the dns resolution of this node and the about to be built manager vm - make sure the dns resolves correctly to the systems you're about to build.
*
hosted-engine --deploy --4
*
Pretty much hit enter for most things, except for the following:
*
Please specify which way the network connectivity should be checked (ping, dns, tcp, none) [dns]:ping
*
Engine VM FQDN:emanager.<fqdn>
*
Enter root password that will be used for the engine appliance:<your password>
*
How should the engine VM network be configured? (DHCP, Static)[DHCP]:Static
*
Please enter the IP address to be used for the engine VM []:192.168.1.251 # the ip address emanager.<fqdn> resolves to.
*
Enter engine admin password:<your password>
*
Please provide the hostname of this host on the management network [ovnode01]:ovnode01.<fqdn> # fully qualified name that we know resolves to the correct ip address.
*
....... Lots of stuff happens
*
[ INFO ] TASK [ovirt.ovirt.hosted_engine_setup : Install oVirt Hosted Engine packages]
[ INFO ] ok: [localhost]
...
*
[ INFO ] TASK [ovirt.ovirt.hosted_engine_setup : Install ovirt-engine-appliance rpm] (sits here for quite a while as the rpm contains the appliance vm, it has to install the rpm, unwind it, copy bits and pieces around etc)
*
Eventually you'll get to
*
Please specify the storage you would like to use (glusterfs, iscsi, fc, nfs)[nfs]:nfs
*
Please specify the nfs version you would like to use (auto, v3, v4, v4_0, v4_1, v4_2)[auto]:
*
Please specify the full shared storage connection path to use (example: host:/path):192.168.1.250:/ovirtx
*
If needed, specify additional mount options for the connection to the hosted-engine storagedomain
*
(example: rsize=32768,wsize=32768) []:
...
Please specify the size of the VM disk in GiB: [51]:
Building engine vm and shoving it out on NFS (In my case)
...
[ INFO ] TASK [ovirt.ovirt.hosted_engine_setup : Shutdown local VM]
[ INFO ] changed: [localhost]
[ INFO ] TASK [ovirt.ovirt.hosted_engine_setup : Wait for local VM shutdown]
...
TASK [ovirt.ovirt.hosted_engine_setup : Copy local VM disk to shared storage]
...
[ INFO ] Stage: Pre-termination
[ INFO ] Stage: Termination
[ INFO ] Hosted Engine successfully deployed
And hey presto the manager should be at https://emanager.<fqdn>
________________________________
From: Marco Fiorletta
Sent: Monday, March 03, 2025 7:00 PM
To: Geoff O'Callaghan; users(a)ovirt.org
Subject: Re: [ovirt-users] Re: Couldn't resolve host name for http://mirrorlist.centos.org/?release=8-stream&arch=x86_64&repo=storage-c...
Hey there, I've been running into a really frustrating issue while trying to set up a self-hosted engine. I've tried multiple times, with different distributions like CentOS Stream 9, AlmaLinux 9.5, and Rocky Linux 9, but I keep hitting the same wall.
Essentially, during the engine configuration, I get an error saying that https://mirroros.centos.org is unreachable.
I've tried all the repository workarounds I've seen online, changing the pointers and everything, but nothing seems to help.
The weird thing is, if I install the engine on a dedicated server, it works fine. It's only when I try to deploy a self-hosted engine that I get this mirror error.
Then, just to see what would happen, I gave Oracle Linux a shot. And guess what?
The self-hosted engine configured perfectly, without even trying to touch those CentOS repositories.
So, that's definitely pointing to some kind of distribution-specific issue.
On top of that, I've also run into another problem with the 4.5 distribution related to VLANs.
I'm working in a data center with a bunch of VLAN-certified networks.
I wanted to set up oVirt 4.5, using a self-hosted engine, and assign one of those VLAN networks to ovirtmgmt.
But the installation just flat-out fails because it only seems to accept physical network devices.
The only way I could get it to work was to put a switch in place to mask the VLAN, which is a bit of a clunky workaround.
My older 4.4-based virtualizer didn't have any of these problems, so I'm wondering if something's changed significantly in 4.5. Any ideas?
Best regards
On Sun, 2025-03-02 at 20:56 +0000, Geoff O'Callaghan wrote:
Hi
I'm not sure where things are going wrong for you, so I built a new Centos 9 Stream system and was able to install a self-hosted appliance on it - ie. It got turned into a node and the engine appliance was deployed OK. I just wanted to make sure the current process worked.
Can you provide more information on what/how you're doing the installation? It might be that the documentation is wrong, but I will fix it if you can tell me where you're going wrong.
Thanks
Geoff
________________________________
From: fiorletta--- via Users
Sent: Tuesday, February 25, 2025 3:30 AM
To: users(a)ovirt.org
Subject: [ovirt-users] Re: Couldn't resolve host name for http://mirrorlist.centos.org/?release=8-stream&arch=x86_64&repo=storage-c...
I have the same problem.
Centos stream 9, ovirt 4.5 the errore is present only on self-hosted-engine.
_______________________________________________
Users mailing list -- users(a)ovirt.org
To unsubscribe send an email to users-leave(a)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/5TXJXB5FCV3...
--
[cid:d80ea0d0d3efc5e30a5e8dbbdb04362ac24caae7.camel@ssolo.eu-0]
1 week, 5 days
