The reply button is making me write an email, so hopefully this reaches you
all.
I think this thread (even if the title is not so clear about the
discussions born inside) could be a good read regarding single host
limitations in terms of updating the environment, after the initial
deployment
Cool, thanks for the advice! Based on the link, it seems like the hosted
engine can't update on a "single machine" install. But if you install the
engine directly to the host, the VM doesn't need to be running, so it might
be able to. Thankfully I wasn't too far down playing with the hosted
engine, or container setup yet.
So with installing an engine directly to the node, I got it to install
ovirt-engine, by enabling the appstream, baseos, extras, and powertools
repo. Then I commented out all "includepkgs = ..." in both
`/etc/yum.repos.d/ovirt-*.repo` files. I can also get through all of the
`engine-setup` questions, but then it fails with starting the
`ovirt-imageio` service at the very end.
The logs in `/var/log/ovirt-imageio/daemon.log`:
```txt
File "/usr/lib64/python3.6/site-packages/ovirt_imageio/_internal/ssl.py",
line 16, in server_context
purpose=ssl.Purpose.CLIENT_AUTH, cafile=cafile)
File "/usr/lib64/python3.6/ssl.py", line 468, in create_default_context
context.load_verify_locations(cafile, capath, cadata)
FileNotFoundError: [Errno 2] No such file or directory
```
Which is the same error as this forum here:
https://lists.ovirt.org/archives/list/users@ovirt.org/thread/DNB73ZMUB5DI...
But all the info about solving it is " Found it; Default route was set to
the NFS / SAN Network Gateway, not the Actual gateway.", and I'm not
exactly sure what that means.
I also tried the engine install/setup in a CentOS workstation VM, and it
worked great, so I thought the node is missing a package? I installed
`openssl-devel` without any luck. In both the VM and the node, I tried
`route -n`, and both CentOS vm / node had similar/expected output. I tried
to do as many default options as possible on both, so I'm surprised the
gateway is different between them.
Any idea how to get past this? I'm also open to switching to plain CentOS
if that might be more stable, and installing to that. I just love how a lot
of the node comes pre-packaged for you, and I'm not sure if you can install
a "node hypervisor" straight to CentOS. That's also outside my comfort
zone, so I'm open to advice either way.
Thanks all for letting me get just this far!