On Mon, Oct 10, 2016 at 9:22 PM, Susinthiran Sithamparanathan <chesusin@gmail.com> wrote:
On Mon, Oct 10, 2016 at 11:40 AM, Simone Tiraboschi <stirabos@redhat.com> wrote:

OK, the issue on the host is just here:
MainThread::DEBUG::2016-10-10 11:18:34,169::brokerlink::282::ovirt_hosted_engine_ha.lib.brokerlink.BrokerLink::(_communicate) Full response: success {"reason": "failed liveliness check", "health": "bad", "vm": "up", "detail": "up"}
MainThread::DEBUG::2016-10-10 11:18:34,169::brokerlink::255::ovirt_hosted_engine_ha.lib.brokerlink.BrokerLink::(_checked_communicate) Successful response from socket
MainThread::DEBUG::2016-10-10 11:18:34,170::brokerlink::151::ovirt_hosted_engine_ha.lib.brokerlink.BrokerLink::(get_monitor_status) Success, status {"reason": "failed liveliness check", "health": "bad", "vm": "up", "detail": "up"}

the engine VM goes up but the engine no and so after a certain amount of time it tries again with a reboot.
We should definitively add a more explicit log entry there!

Now the point is just why your engine is not starting.
Perhaps you ignored https://paste.fedoraproject.org/447579/14760912/? It seems the reason it doesn't start is due to some missing SSL/TLS certs inside the VM. Or do you think the problem is still somewhere else?

No, the issue is definitively there but I was trying to identify the root cause.
 
I've uploaded engine-setup logs to https://my.owndrive.com/index.php/s/3Dcyho9bqo7oZs8 in the folder engine-setup and you'll see many.

You uploaded ovirt-hosted-engine-setup logs from the host while now we need engine-setup logs from the engine VM.
 
I did desperately try to get up the HE when it failed after upgrade by trying to manually setup a new HE which failed (couple of times IIRC). So that might have caused couple of additional logs files in there.


--

Susinthiran Sithamparanathan