Sorry, I guess I skipped a bit of that.
I did run hosted-engine --deploy and it failed while waiting for vdsm to
respond and at that point I sent the original email because I wasn't quite
certain if it was a supported to migrate from non-hosted engine to a hosted
engine which seems to be possible now (thanks for that link). What I had done
was leave that engine in place, try to install the hosted engine on a host
already part of the original non-hosted engine configuration which seems to be
a bad and somewhat futile path. It looks like the best option for me now if I
choose to go to a hosted-engine is to use the migration instructions you sent
and do that, leave it as it is or stand up a seperate cluster and start from
On Tue, Sep 20 , 2016 at 4:40 PM, Logan Kuhn < log...@wolfram.com > wrote:
| It's still running on a physical machine. I saw that article, is the way
| to migrate my physical engine to a hosted engine and then I can deploy more
| hosted engines?
In hosted-engine scenario, the engine is running on a VM hosted by one of the
hosts it's managing.
The normal procedure is to run hosted-engine-setup on the first host and it
will create a VM for you with the engine; hosted-engine-setup will also add
that host to your engine when available.
For who deployed in the past we also have a migration procedure.
Now I don't understand if you still have to try to migrate to hosted-engine or
if you tried it and something went bad.
Did you have ever tried to run 'hosted-engine --deploy' on one host?
Did it completed successfully?
| ----- On Sep 20 , 2016, at 9:36 AM, Simone Tiraboschi < stira...@redhat.com >
|| On Tue, Sep 20 , 2016 at 4:28 PM, Logan Kuhn < log...@wolfram.com > wrote:
||| I tried this and I get the attached error message.
||| Error while executing action: Cannot add Host. You are using an unmanaged
||| engine VM. Please upgrade the cluster level to 3.6 and wait for the hosted
||| engine storage domain to be properly imported.
||| I used only the required information. Power management is not configured,
||| although even with power management configured (ipmilan) it gives the same
||| error message.
||| The cluster and data center compatibility levels are at 4.0.
||| I also created a new data center and cluster and the problem continues. This
||| installation was done via engine-setup and is on a physical machine, not a
||| hosted-engine itself. Is that the problem?
|| Sorry, but do you mean that you initially deployed the engine on a physical
|| machine and then you migrated your setup to an hosted-engine scenario
||  or is your engine still running on a physical machine?
||  -
||| ----- On Sep 19 , 2016, at 3:06 PM, Simone Tiraboschi < stira...@redhat.com
|||| On Mon, Sep 19 , 2016 at 9:53 PM, Logan Kuhn < log...@wolfram.com > wrote:
||||| I understand and I would prefer that as well, but for the life of me I
||||| seem to find where that is available in the web admin.
|||| See the attached picture.
||||| ----- On Sep 19 , 2016, at 2:44 PM, Simone Tiraboschi <
|||||| On Mon, Sep 19 , 2016 at 6:59 PM, Logan Kuhn < log...@wolfram.com >
||||||| When I try to deploy a second hosted engine it errors out at
||||||| Full log attached.
||||||| The host I'm deploying to is actually going to end up being our
||||||| instance of the image. I've also tried to install it on a currently
||||||| VM host with the exact same error message.
|||||| Hi Logan,
|||||| could you please try with
|||||| from 4.0.3 RC3?
|||||| Patch https://gerrit.ovirt.org/#/c/63001/ should also address your issue.
|||||| In general with 4.0 I'd strongly suggest to deploy additional
|||||| hosts directly from the engine web admin instead of using
|||||| also there.
||||||| Users mailing list
Users mailing list