On Sat, May 14, 2022 at 12:53 AM <louisb(a)ameritech.net> wrote:
I’ve tried your suggestion and continue to get the same results.
It will be more helpful if you describe in detail what you tried
and what the results were.
Based on continuing investigation I’ve found on the Red Hat Knowledge
base a resolution to this issue, the following link references the solution:
https://access.redhat.com/solutions/354255.
This URL does not exist.
However, I’ve run across another issue, since the creation of a new
host within ovirt
Based on the output of "ovirt-imageio --show-config" you had an all-in-one
setup, when the engine host is added to the engine as a hypervisor. This setup
is not supported but works, but adding more hosts to this kind of setup will not
work with image transfer, and it is a really bad idea to have multiple hosts and
run engine on one of them.
For example, engine can stop a host using host power management API. If this
is the host running engine you don't have a way to start your engine unless you
have access to the host power management console.
If have more than one host, your engine should not run on any of the hosts,
and you must enable the imageio proxy (this is the default):
engine-config -s ImageTransferProxyEnabled=true
And restart engine:
systemctl restart ovirt-engine
I’ve not been able to access the internet or reach the host/server
remotely. Therefore, I’m unable to try the solution provide via the Red Hat Knowledge
Base.
I’ve reviewed the kernel routing table displayed below:
ip route show
default via 20.10.20.1 dev eno5 proto static metric 100
default via 20.10.20.1 dev eno6 proto static metric 101
default via 20.10.20.1 dev eno7 proto static metric 102
default via 20.10.20.1 dev ovirtmgmt proto static metric 425
20.10.20.0/24 dev eno5 proto kernel scope link src 20.10.20.65 metric 100
20.10.20.0/24 dev eno6 proto kernel scope link src 20.10.20.66 metric 101
20.10.20.0/24 dev eno7 proto kernel scope link src 20.10.20.67 metric 102
20.10.20.0/24 dev ovirtmgmt proto kernel scope link src 20.10.20.68 metric 425
Is it normal behavior for the host to sever all connection when a “Host” machine is added
to ovirt? Is there a solution to this issue? I’ve recognize the risks of having the
host exposed to the internet, how would I keep the OS/RHEL 8.6 & ovirt current?
It is hard to tell what's going on when we don't know which
hosts do you have and what is their ip address.
Please confirm that you access engine using https:// and that
when you access the host your browser reports a secure connection
without warnings (meaning that engine CA certificate was added to
the browser).
Nir