Hey! Have you added the hostname to the known_hosts file or setup passwordless ssh for the single node? Ansible requires a passwordless ssh to ensure root access to automate the installation process. If yes, then probably this issue is fixed in the next build of cockpit-ovirt. Can you try to upgrade and check again? On Wed, Aug 19, 2020 at 6:32 AM David White via Users <users(a)ovirt.org&gt; wrote:

Thank you. I read through the "ovirt-imageio: can't upload / download" thread, and your brief glossary was very helpful. Perhaps it would make sense to put some basic terminology like that somewhere in the official oVirt docs? I feel like there are a million different ways to install and configure oVirt, and it has taken me quite a bit of time to figure out the differences. I took your advice, and figured out how to get the "standalone engine" installed and running. I'm now able to log into the oVirt Admin Portal, view / create Datacenters, etc.. At the risk of creating a redundant thread, I'm going to reply to this thread about a problem I'm having uploading ISOs through the Admin Portal. I can't do it! Perhaps my issue is related to the same issue Michael is facing in the thread I just referenced - ovirt-imageio: can't upload/download. I have the standalone install: [dwhite@dev1-centos ~]$ sudo yum info ovirt-engine | grep Version Version : 4.4.1.10 [dwhite@dev1-centos ~]$ sudo systemctl status ovirt-imageio ● ovirt-imageio.service - oVirt ImageIO Daemon Loaded: loaded (/usr/lib/systemd/system/ovirt-imageio.service; enabled; vendor preset: disabled) Active: active (running) since Fri 2020-08-21 19:16:28 EDT; 49min ago Main PID: 1435 (ovirt-imageio) Tasks: 3 (limit: 409555) Memory: 18.8M CGroup: /system.slice/ovirt-imageio.service └─1435 /usr/libexec/platform-python -s /usr/bin/ovirt-imageio Aug 21 19:16:28 dev1-centos.office.barredowlweb.com systemd[1]: Starting oVirt ImageIO Daemon... Aug 21 19:16:28 dev1-centos.office.barredowlweb.com systemd[1]: Started oVirt ImageIO Daemon. This is of course a new, default install. I have created a data directory for VM images, as well as ISOs: [dwhite@dev1-centos ~]$ ls -la /data/ total 0 drwxr-xr-x. 4 vdsm kvm 31 Aug 19 20:49 . dr-xr-xr-x. 19 root root 260 Aug 19 20:28 .. drwxr-xr-x. 3 vdsm kvm 50 Aug 21 20:12 images drwxr-xr-x. 3 vdsm kvm 50 Aug 21 20:12 iso I am now trying to upload an ISO through the Admin Portal by going to Storage -> Disks -> Upload -> Start. My client machine is Ubuntu. I click "Test Connection" and I get the error message: Connection to ovirt-imageio-proxy service has failed. Make sure the service is installed, configured, and ovirt-engine certificate is registered as a valid CA in the browser. I downloaded the certificate by clicking on the link provided, and saved it into /usr/share/ca-certificates/mozilla/ I then ran: sudo update-ca-certificates Then I reloaded the oVirt admin portal. I'm still getting the error message. Any advice here? Or am I running into the same problem that Michael has run into?

On 22/08/2020 11:20, David White via Users wrote: 1. download it, it will save as pki-resource 2. rename file to blah.pem 3. open firefox > preferences > privacy and security > view certificates 4. open authorities tab 5. import > select blah.pem 6. check the box for trust to identify websites make sure to remove any exception granted, so it uses ca instead of exception for auth.

I see you are using 4.4 standalone; by standalone do you mean all-in-one? or separate engine and hosts? In 4.4 the change in imageio means it's not compatible with all-in-one anymore, i'll be raising a bug request to support this later today (but officially all-in-one support ended in 4.0 so i don't know if the bug will be actioned). Kind Regards, Mike On 22/08/2020 12:06, David White via Users wrote:

On 22/08/2020 13:58, David White via Users wrote: You can actually upload directly via scp to the iso domain, just make sure to; chown -v vdsm: /path/to/iso/domain/local_iso_domain/0000000-0000-0000-0000-000000/images/11111111-1111-1111-1111-111111111111/uploaded-file.iso the system will still see it and you can attach to vms at boot; you can also use pxe, i like the "fai" project for this. 4.4 all-in-one is still fully functional par the iso upload and download, which i was using for backups. if you were to go CentOS7+4.3 these features fully work no problem in all-in-one. the alternate to all-in-one would be to choose a different host when adding to the standalone manager, or you can have hosted engine, where the engine is a vm (dependent on some other stuff re: storage/ips) Kind Regards, Mike

On 22/08/2020 18:20, David White via Users wrote: when you create the iso domain, it should have created the path for you, 0000-000-00 was just an example path; for local iso domain, it should be along the lines of; 1. create top level dir and chown to vdsm: in your case "/data/iso" 2. add iso domain in ui 3. when it added the local iso domain it will setup a bunch of subfolders, in the end folder which is 11111-111-111 you can dump all your iso images in that 1x dir... ie, move the iso file "ubuntu-20.04.1-live-server-amd64.iso" into the same dir as "CentOS-8.2.2004-x86_64-minimal.iso" check the engine logs yum update will only ensure the packages are installed, however this doesn't mean ovirt has run the upgrade, just upgrade through the ui; 1. put host to maint 2. update host this will trigger ansible which; 1. does yum updates 2. updates the ovirt version to match your latest package you can check what the upgrade is doing as there is a one off log file each time you click update. can't remember the path off the top of my head, just look in /var/log/ovirt*

On Sat, Aug 22, 2020 at 3:19 AM David White <dmwhite823(a)protonmail.com&gt; wrote: I agree this makes sense. Now searched and failed to find such a page. Please open a doc bug to add it. Thanks! There is such a page for RHV, which should be around 99% applicable, which you can read for the time being: https://access.redhat.com/documentation/en-us/red_hat_virtualization/4.3/... Best regards, -- Didi