Hi again,

I manage to go a little bit further.. I was not able to set one host to maintenance because they had running vm.. so i force it to mark it as reboot  and flush any vm and now i can try to reinstall the host.. but now i am getting error when the installation try to enroll certificate..

Any idea?

Thanks
Carl


On Sat, Jan 6, 2018 at 9:41 AM, carl langlois <crl.langlois@gmail.com> wrote:
Hi,

Thanks for the quick reply.

Version before the update was the latest 4.1. 
The storage is NFS.
The host are plain 7.4 Centos.i keep them up to date
if i remenber the initail deploy of the engine vm was full OS

The step to upgrade was:

SSH to hosted engine vm , 
yum install 4.2 release
run engine-setup.
this where i think something when wrong.
the engine setup was not able to update some certificate.. a quick search on the internet and found that moving the ca.pem from the /etc/pki/ovrit-engine made the engine-setup work..

from that point the hosted_engine seem to work fine but all hosts and data center are not operational.

One of my main concern is losing all the template and users vm.. 



Engine and vdsm log of one host are here..

In the engine log there is a lot of this error..

2018-01-06 08:31:40,459-05 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.GetCapabilitiesVDSCommand] (EE-ManagedThreadFactory-engineScheduled-Thread-9) [] Command 'GetCapabilitiesVDSCommand(HostName = hosted_engine_2, VdsIdAndVdsVDSCommandParametersBase:{hostId='7d2b3f49-0fbb-493b-8f3c-5283566e830d', vds='Host[hosted_engine_2,7d2b3f49-0fbb-493b-8f3c-5283566e830d]'})' execution failed: VDSGenericException: VDSNetworkException: General SSLEngine problem.

https://drive.google.com/drive/folders/1tZHLIMV0ctGyeDPcGMlJa6evk6fkrCz8?usp=sharing

Thanks for your support

Carl

 

On Sat, Jan 6, 2018 at 8:50 AM, Gianluca Cecchi <gianluca.cecchi@gmail.com> wrote:
On Sat, Jan 6, 2018 at 2:42 PM, carl langlois <crl.langlois@gmail.com> wrote:
Hi again all,

I really need your input on recovering my failed upgrade to 4.2.



First two things to do to get appropriate help would be, in my opinion:

1) provide a detailed description of your environment
exact version before the update; kind of storage (Gluster, NFS, iSCSI, FC), kind of hosts (plain OS and which version or node-ng and which version before and after), kind of deploy executed for hosted engine VM at initial install time (appliance, full OS).
Describe steps used for upgrading.

2) upload to a shared file services all the possible logs:
engine logs before and after the upgrade and logs of engine-setup run during update
vdsm logs before and after the upgrade


Gianluca