12:51 a.m.
Hi,
You can follow the steps below to do that.
1) Stop all the virtual machines.
2) Move all the storage domains other than hosted_storage to maintenance
which will unmount them from all the nodes.
3) Move HE to global maintenance 'hosted-engine --set-maintenance --mode
=global'
4) stop HE vm by running the command 'hosted-engine --vm-shutdown'
5) confirm that engine is down using the command 'hosted-engine --vm-status'
6) stop ha agent and broker services on all the nodes by running the
command 'systemctl stop ovirt-ha-broker' ; 'systemctl stop
ovirt-ha-agent'
7) umount hosted-engine from all the hypervisors 'hosted-engine
--disconnect-storage'
8) stop all the volumes.
9) power off all the hypervisors.
To bring it up back again below steps will help.
1) Power on all the hypervisors.
2) start all the volumes
3) start ha agent and broker services on all the nodes by running the
command 'systemctl start ovirt-ha-broker' ; 'systemctl start
ovirt-ha-agent'
4) Move hosted-engine out of global maintenance by running the command
hosted-engine
--set-maintenance --mode =none
5) give some time for the HE to come up. check for hosted-engine
--vm-status to see if HE vm is up.
6) Activate all storage domains from UI.
7) start all virtual machines.
Hope this helps !!!
Thanks
kasturi.
On Tue, Aug 8, 2017 at 2:27 AM, Moacir Ferreira <moacirferreira(a)hotmail.com>
wrote:
Sorry Erekle, I am just a beginner...
From the hosted engine I can put the two other servers, that are not
hosting the hosted-engine, on maintenance, and that was what I did. When I
tried to put the last one on maintenance it did not allow me due to the
hosted-engine and I force it shunting down from the ssh CLI.
So, what should I do? My guess is that I should ssh to the hosted engine
and shut it down. As it would not have another node to re-launch itself,
then it would stay down. After this I should shutdown the oVirt node. Is it?
Anyway, I made a mistake and I forced it. When I try to bring the cluster
back, I noticed that glusterd was not enabled to start when the node
power-on. As I am a beginner, I "think" that there must be a reason to not
start glusterd when the node comes up. So I started glusterd on the
arbitrator (3rd server), then on the second node and finally on the host
that was hosting the hosted-engine. It worked, but when I try to change the
maintenance mode on the two nodes (2nd and 3rd) back to normal, the
hosted-engine went down and I had to start it manually.
All this said, I would like to know how to bring the cluster down and how
to bring it back on in the "right way" so I don't get problems. And yes,
no
VM is running but the hosted-engine.
Thanks for sharing your knowledge.
Moacir
------------------------------
*From:* Erekle Magradze <erekle.magradze(a)recogizer.de>
*Sent:* Monday, August 7, 2017 9:12 PM
*To:* Moacir Ferreira; users(a)ovirt.org
*Subject:* Re: [ovirt-users] How to shutdown an oVirt cluster with
Gluster and hosted engine
Hi Moacir,
First switch off all VMs.
Second you need to declare hosts maintenance mode, don't start with SRM
(of course if you are able use the ovirt-engine), it will ask you to
shutdown the glusterfs on a machine.
Third if all machines are in maintenance mode, you can start shutting down
them.
If you have hosted engine setup follow this [1]
Cheers
Erekle
[1] https://github.com/rharmonson/richtech/wiki/OSVDC-Series:-
oVirt-3.6-Cluster-Shutdown-and-Startup
<https://github.com/rharmonson/richtech/wiki/OSVDC-Series:-oVirt-3.6-Clust...
OSVDC Series: oVirt 3.6 Cluster Shutdown and Startup ...
<https://github.com/rharmonson/richtech/wiki/OSVDC-Series:-oVirt-3.6-Clust...
github.com
Contribute to richtech development by creating an account on GitHub.
On 08/07/2017 08:58 PM, Moacir Ferreira wrote:
I have installed a oVirt cluster in a KVM virtualized test environment.
Now, how do I properly shutdown the oVirt cluster, with Gluster and the
hosted engine?
I.e.: I want to install a cluster of 3 servers and then send it to a
remote office. How do I do it properly? I noticed that glusterd is not
enabled to start automatically. And how do I deal with the hosted engine?
Thanks,
Moacir
_______________________________________________
Users mailing listUsers@ovirt.orghttp://lists.ovirt.org/mailman/listinfo/users
--
Recogizer Group GmbH
Dr.rer.nat. Erekle Magradze
Lead Big Data Engineering & DevOps
Rheinwerkallee 2, 53227 Bonn
Tel: +49 228 29974555
E-Mail erekle.magradze(a)recogizer.de
Web: www.recogizer.com
Recogizer auf LinkedIn https://www.linkedin.com/company-beta/10039182/
Folgen Sie uns auf Twitter https://twitter.com/recogizer
-----------------------------------------------------------------
Recogizer Group GmbH
Geschäftsführer: Oliver Habisch, Carsten Kreutze
Handelsregister: Amtsgericht Bonn HRB 20724
Sitz der Gesellschaft: Bonn; USt-ID-Nr.: DE294195993
Diese E-Mail enthält vertrauliche und/oder rechtlich geschützte Informationen.
Wenn Sie nicht der richtige Adressat sind oder diese E-Mail irrtümlich erhalten haben,
informieren Sie bitte sofort den Absender und löschen Sie diese Mail.
Das unerlaubte Kopieren sowie die unbefugte Weitergabe dieser Mail und der darin
enthaltenen Informationen ist nicht gestattet.
_______________________________________________
Users mailing list
Users(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/users