On Wed, Jun 17, 2020 at 10:12 AM Yedidyah Bar David <didi(a)redhat.com> wrote:
Section 5.3 is not actually needed, the appliance already includes
it.
> The second document is missing the modules part.
(Seems to me to include it as well, but again, it's not needed)
> Now I can manually install ovirt-engine on the host.
I think you got it wrong. Please see my previous posts in this thread.
In particular:
1. You do not need to install ovirt-engine on the host.
2. You should try to see why it failed, on the engine VM.
3. This engine vm should be left running by the failed deploy process,
and you can find its IP address by searching the setup logs for
'local_vm_ip'. So please find it, ssh to it from the host (using the
root password you supplied), then try 'dnf install ovirt-engine'. If
it fails, try to fix the problem. My guess: Some kind of
networking/mirror/proxy/whatever issue.
For an overview of the current hosted-engine deploy process, you can
have a look at "Hosted Engine 4.3 Deep Dive (Simone Tiraboschi)" in:
https://www.ovirt.org/community/get-involved/resources/slide-decks.html
It was written for 4.3, but most of it applies also for 4.4.
Also:
The engine package is already part of the appliance, so strictly
speaking, we should not try to install it. This is done so that if a
newer version is available (e.g. because you installed an older
appliance image rpm), you upgrade to the newer version.
We have a bug open for allowing to prevent this, not sure about its
current status - see also the discussion in the linked github issue:
https://bugzilla.redhat.com/show_bug.cgi?id=1816619
Hello,
My mistake. I think there is a miss-understanding on my end that
side-stepped the main issue.
Before I begin, I'd like to point out that I have a number of oVirt
4.3 deployments in production.
Some installed via cockpit, others manually via hosted-engine --deploy.
I'm attempting to cleanly test-install 4.4 on a CentOS 8.x machine
before I begin (slowly) upgrading my existing 4.3 setups to 4.4.
However, when trying to install 4.4 on the test CentOS 8.x (now 8.2
after yesterday release), either manually (via hosted-engine --deploy)
or by using cockpit, fails when trying to download packages (see
attached logs) during the hosted engine deployment phase.
The reason I mocked around with the java module (on the host) was in
an attempt to see if the _host_ machine somehow had the wrong module
configuration, triggering the hosted engine deployment process.
Just to be clear, it is the hosted engine VM (during the deployment
process) that fails to automatically download packages, _not_ the
host.
Sorry again for the misunderstanding.
- Gilboa