- Users - oVirt List Archives

Impossibile to install a new system
by Andrea Chierici 22 Sep '23

22 Sep '23

Dear all, I've been struggling with a new ovirt installation for several weeks but I seem to have no luck at all. I've tried both starting from a centos8 stream system and a ovirt-node, but the engine installation always breaks here: TASK [ovirt.ovirt.hosted_engine_setup : Wait for the host to be up] I keep getting the same problem, it looks like the engine is not "migrated" to the public ip and is stuck on the natted address that the installation process is using during the preliminary phase. Here is a message that seems to confirm my theory: 2023-09-19 15:43:29,259+0200 DEBUG otopi.ovirt_hosted_engine_setup.ansible_utils ansible_utils._process_output:109 {'msg': "The task includes an option with an undefined variable. The error was: 'local_vm_ip' is undefined\n\nThe error appears to be in '/usr/share/ansible/collections/ansible_collections/ovirt/ovirt/roles/hosted_engine_setup/tasks/sync_on_engine_machine.yml': line 2, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n---\n- name: Set the name for add_host\n ^ here\n", '_ansible_no_log': False} 2 I know the problem the development of ovirt is facing lately, but I am simply wondering: is there a way to install ovirt 4.5.4 today or is it all already lost??! Any kind of help would be really appreciated. Thanks, Andrea -- Andrea Chierici - INFN-CNAF Viale Berti Pichat 6/2, 40127 BOLOGNA Office Tel: +39 051 2095463 SkypeID ataruz --

3 4

ovirt self-hosted engine 4.5 with cluster version 4.6
by Николаев Алексей 20 Sep '23

20 Sep '23

1 0

In-place upgrade of Windows OS
by Hean-Seng Tan 19 Sep '23

19 Sep '23

We are unable to do an in-place upgrade of the Windows OS for a vm from Windows Server 2012r2 to version 2016. Has anyone ever performed successfully an in-place upgrade? Can you please share the steps?

1 0

oVirt 4.5.4-1: engine-setup does not renew certificates
by Lars Stolpe 15 Sep '23

15 Sep '23

Hi, I wanted to add short names to the webgui apache certificate. I followed these instructions: „Engine can be accessed using alternate host names (or IP addresses) that can configured by adding a new configuration file (for example /etc/ovirt-engine/engine.conf.d/99-custom-sso-setup.conf) with following content: SSO_ALTERNATE_ENGINE_FQDNS="alias1.example.com alias2.example.com" The list of alternate names has to be listed separated by spaces. It's possible to add also IP addresses of engine host, but using IP addresses instead of DNS names is not considered to be a good practise. Run engine-setup to generate new certificates.“ Certificates where not renewed. How do i get engine-setup to renew the apache certificate? oVirt: 4.5.4-1 Regards, Lars

2 2

How to turn off this checks?
by Jorge Visentini 13 Sep '23

13 Sep '23

Hi guys, Is it possible to disable these checks? *1 - Failed to verify Power Management configuration for Host ovirtHost.domain.* *2 - Available swap memory of host ovirtHost.domain [2 MB] is under defined threshold [1024 MB].* I know that is not a good practice, but I don't need it for now. Thanks all! -- Att, Jorge Visentini +55 55 98432-9868

1 0

Version 4.5. How should I properly replace a failed host?
by ziyi Liu 12 Sep '23

12 Sep '23

gluster uses 2+1 arbitration mode My host is damaged and I need to reinstall the system. The host fqdn still uses the old fqdn. What should I do? I found some historical information. Is it still usable for the new version? Found gluster-ansible in GitHub https://github.com/gluster/gluster-ansible/blob/master/playbooks/hc-ansible… See the instructions and configure node_prep_inventory.yml and node_replace_inventory.yml to restore. Execute ansible-playbook -i node_prep_inventory.yml -i node_replace_inventory.yml tasks/replace_node.yml My data disk /dev/sdb has also been initialized by me. Can this ansible script help me create a new partition, create a mount, and complete the recovery of the gluster node?

1 0

ssl stomp reactor timeout
by Chris Smith 12 Sep '23

12 Sep '23

Hi all, I'm using ovirt 4.2 so I'm a little behind, but I'm trying to understand what this error message means. ERROR [org.ovirt.vdsm.jsonrpc.client.reactors.ReactorClient] (SSL Stomp Reactor) [] Connection timeout for host 'x.x.x.x', last response arrived 1549 ms ago. Is this initiated by the engine server or the ovirt host ? Is there a way for me to replicate what ovirt is doing manually via a telnet or curl type command ? Thanks, -- Chris -- Squiz named market leader (second year) in Omdia's Selecting a Digital Experience Management Solution report, 2022-23. Download the report today! <https://www.squiz.net/omdia-report-2022-23?utm_source=Website&utm_medium=Do…>

1 0

engine python sdk ImportError: cannot import name 'Client'
by parallax 11 Sep '23

11 Sep '23

I try to use the Python sdk to download vm disk but get the error: # python3.9 download_disk.py f556403f-2a32-4e35-9f17-dc67864e0ae1 /home/librenms.raw --format raw --config engine2 Traceback (most recent call last): File "download_disk.py", line 31, in <module> from ovirt_imageio import client ModuleNotFoundError: No module named 'ovirt_imageio' or when run it with python3.6: # python3.6 download_disk.py f556403f-2a32-4e35-9f17-dc67864e0ae1 /home/librenms.raw --format raw --config engine2 Traceback (most recent call last): File "download_disk.py", line 31, in <module> from ovirt_imageio import client ImportError: cannot import name 'client' Environment: Rocky Linux release 8.8 Python 3.6.8, Python 3.9.16 python39-ovirt-engine-sdk4-4.6.0-1.el8.x86_64, python3-ovirt-engine-sdk4-4.6.2-1.el8.x86_64 ovirt-release44-4.4.10.3-1.el8.noarch

1 1

Testing oVirt backup and Restore operations 4.3
by Mikhail Po 07 Sep '23

07 Sep '23

There was a desire to conduct a test backup and restore operation ovirt-node-ng 4.3.2-2019031908 Operations performed: - Deployed two test hosts, configured Hosted Engine. Created two test VMs. - Created a backup copy # engine-backup --mode=backup --file=ovirt-full-backup_2023-09-04.tar --log=ovirt-full-backup_2023-09-04.log - Turned off the original Self-hosted Engine and disconnected from Self-hosted Domain Storage # hosted-engine --vm-shutdown # hosted-engine --disconnect-storage - Created a new host using a previously created backup # hosted-engine --deploy --restore-from-file=ovirt-full-backup_2023-09-04.tar Although an error occurred at the end of the installation (after the start of copying to the iSCSI shared storage), but oVirt is accessible via the Web. The error that occurred [ INFO ] TASK [ovirt.hosted_engine_setup : Check for the local bootstrap VM] [ ERROR ] Error: The response content type 'text/html; charset=iso-8859-1' isn't the expected XML [ ERROR ] fatal: [localhost]: FAILED! => {"changed": false, "msg": "The response content type 'text/html; charset=iso-8859-1' isn't the expected XML"} [ ERROR ] Failed to execute stage 'Closing up': Failed executing ansible-playbook ... [ INFO ] Stage: Clean up [ INFO ] Cleaning temporary resources [ ERROR ] Hosted Engine deployment failed: please check the logs for the issue, fix accordingly or re-deploy from scratch. Log file is located at /var/log/ovirt-hosted-engine-setup/ovirt-hosted-engine-setup-20230907201859-ovmmky.log In the Web interface - all Hosts are green - all VMs are green. There is a virtual machine "external-HostedEngineLocal". - Storage Domain "hosted_storage_old_20230907T202958" is present. Any questions: 1. The instructions say that at the end you need to reinstall the hosts. Is it necessary to do this? I ask, because in my case, all hosts are available to the manager, virtual machines are working. What is this operation necessary for? 2. What to do with the "external-HostedEngineLocal" virtual machine? 3. Why the "Error" error occurred: The response content type 'text/html; charset=iso-8859-1' isn't the expected XML"? Can it be related to the lack of access to the Internet (repository) have a VM Engine? When installing hosts, access to the Internet is required (otherwise the installation does not occur), I register access through a proxy in yum.conf

1 0

oVirt 4.5.4-1 OVA export generates incomplete XML tag in vm.ovf
by lars.stolpe＠bvg.de 07 Sep '23

07 Sep '23

Hi, we can't import some OVA VMs into oVirt, that was previously exported in oVirt. Further investigations and testing showed, that in some vm.ovf files the last closing tag is missing ( </ovf:Envelope instead of </ovf:Envelope> ). Intense testing showed the following occurrence: on test DC (4.4.10.7-1): - about half of the VMs have that problem (about 9 VMs) - VMs where the export is working properly are ok, no matter how often i do export - VMs who's vm.ovf is wrong stay that way, no matter how often i do export on production DC (4.4.10.7-1): - all VMs have that problem (49 VMs) - an imported "good" VM formerly exported on test DC, now have a wrong vm.ovf too after re-exporting on production DC - vm.ovf of all VMs stay incorrect, no matter how often i do export on develop DC (4.5.4-1): - the only VM have that problem (1 VM) - vm.ovf of VMs stay incorrect, no matter how often i do export Environment: all DC have independent hardware and are built exactly the same way with the same components oVirt is with hosted-engine test and production: oVirt: 4.4.10.7-1 engine OS: Cent OS Stream 8 host OS: RHEL 8.7 develop: oVirt: 4.5.4-1 engine OS: Cent OS Stream 8 (latest updates till September 1st) host OS: RHEL 8.8 (latest updates till September 1st) Is that a known problem? Is there a way to fix this? regards, Lars

2 3