[ovirt-users] Cluster

Yedidyah Bar David didi at redhat.com
Thu Jan 7 12:37:09 UTC 2016 

On Thu, Jan 7, 2016 at 1:18 PM, Michal Skrivanek
<michal.skrivanek at redhat.com> wrote:
>
> On 07 Jan 2016, at 10:56, Budur Nagaraju <nbudoor at gmail.com> wrote:
>
> My Setup looks like this ,
>
> Hosted Engine1 : Installed with  centos 6.7 64Bit CPU sandy Bridge and
> Default  cluster is  in oVirt engine Belongs to  sandy bridge.
> Hosted Engine2 : Installed with  centos 6.7 64Bit CPU sandy Bridge and
> Default  cluster is  in oVirt engine Belongs to  sandy bridge.
>
> oVirt Engine : Installed with Centos 6.7 64 Bit , its hosted in the hosted
> engine1 or 2.
>
> oVirt node: Installed with the version oVirt Node Hypervisor release 3.5
> (0.999.201509102313.el7.centos) belongs to  Intel Nehalem Family
>
>
> assuming that the “Default” cluster with those 2 HE hosts above is 3.6 then
> no
> Or are you talking about a 3.5 setup altogether? and everything is at
> 3.5…then yes. However there might be some caveats due to the special hosted
> engine VM. Didi or someone from integration would need to comment on that

AFAIK you can't migrate the engine vm to a hypervisor (host or node) that is
not a hosted-engine host. you'd need to deploy the node as a hosted-engine
host. ovirt-node allows this too in 3.5+, see e.g.:
http://dougsland.livejournal.com/124084.html

>
>
> so as per your update we can't migrate the oVirt Engine to the node which
> has the version installed 3.5 (0.999.201509102313.el7.centos ?
>
>
>
>
> On Thu, Jan 7, 2016 at 2:40 PM, Michal Skrivanek
> <michal.skrivanek at redhat.com> wrote:
>>
>>
>> On 07 Jan 2016, at 10:06, Budur Nagaraju <nbudoor at gmail.com> wrote:
>>
>> Thanks for the update .
>>
>> Is there a way to migrate the oVirt Engine  to the oVirt node ? as My
>> setup has two hosted engine when I try to migrate Engine  to the ovirt node
>> which is  in different cluster am not able to succeed.
>>
>>
>> for normal VMs - as long as the cluster compatibility is the same you can
>> migrate a VM from your el6-only cluster to a new el7-only cluster - from el6
>> host to el7 host (using the special cross-cluster migration, both clusters
>> need to be set up the same (storage, networks, etc))
>>
>> Note you can do that only one way from el6 to el7, and in cluster levels
>> 3.5 or 3.6.
>>
>> Thanks,
>> michal
>>
>>
>>
>> Getting the same Error.
>>
>> On Thu, Jan 7, 2016 at 2:18 PM, Michal Skrivanek
>> <michal.skrivanek at redhat.com> wrote:
>>>
>>>
>>> > On 07 Jan 2016, at 07:25, Yedidyah Bar David <didi at redhat.com> wrote:
>>> >
>>> > On Thu, Jan 7, 2016 at 12:21 AM, Budur Nagaraju <nbudoor at gmail.com>
>>> > wrote:
>>> >> Can someone help me in the issue ? This is blocking my testing .
>>> >
>>> > The linked engine.log has:
>>> >
>>> > Missing CPU features : model_SandyBridge
>>> >
>>> > Please check/post exact models of servers and their cpus, bios settings
>>> > in both, cluster cpu type. Thanks.
>>>
>>> The log seem to be pretty clear about that, and there are messages even
>>> in UI
>>>
>>> Message: Not possible to mix RHEL 6.x and 7.x hosts in one cluster. Tried
>>> adding oVirt Node - 3.6 - 0.999.201510261021.el7.centos host to a cluster
>>> with RHEL - 6 - 7.el6.centos.12.3 hosts.
>>>
>>> and indeed separate problem with CPU mode, your cluster requests
>>> sandybridge but your host is older
>>>
>>> You need to keep same RHEL versions within cluster, and if you want to
>>> use older HW you need to lower the cluster requirements
>>>
>>> >
>>> >>
>>> >> On Jan 6, 2016 1:57 PM, "Budur Nagaraju" <nbudoor at gmail.com> wrote:
>>> >>>
>>> >>> Hosted engine and and oVirt engine both are having Centos6.x version
>>> >>> I
>>> >>> mean Its (centos6.7) ,while second host am facing that issues.
>>> >>>
>>> >>> On Wed, Jan 6, 2016 at 1:54 PM, Oved Ourfali <oourfali at redhat.com>
>>> >>> wrote:
>>> >>>>
>>> >>>> Adding Sandro, as perhaps it is related to hosted engine in any way.
>>> >>>> Conceptually, you can't mix centos6 and 7 in the same cluster, in
>>> >>>> case
>>> >>>> that's what you did.
>>> >>>>
>>> >>>> On Jan 6, 2016 10:08, "Budur Nagaraju" <nbudoor at gmail.com> wrote:
>>> >>>>>
>>> >>>>> As of now its showing only one hosts that is Hosted Engine and the
>>> >>>>> moment I add second host its failing ,below are the OS version.
>>> >>>>> I have installed Cetos 6.7 in the Hosted engine I don't know why
>>> >>>>> its
>>> >>>>> showing RHEL-6-7 in OS version.
>>> >>>>>
>>> >>>>> OS Version:RHEL - 6 - 7.el6.centos.12.3
>>> >>>>> Kernel Version:2.6.32 - 573.12.1.el6.x86_64
>>> >>>>> KVM Version: 0.12.1.2 - 2.479.el6_7.2
>>> >>>>> LIBVIRT Version: libvirt-0.10.2-54.el6_7.3
>>> >>>>>
>>> >>>>> On Wed, Jan 6, 2016 at 1:34 PM, Budur Nagaraju <nbudoor at gmail.com>
>>> >>>>> wrote:
>>> >>>>>>
>>> >>>>>> As of now its showing only one hosts that is oVirt Engine and the
>>> >>>>>> moment I add second host its failing ,below are the OS version.
>>> >>>>>>
>>> >>>>>> OS Version:RHEL - 6 - 7.el6.centos.12.3
>>> >>>>>> Kernel Version:2.6.32 - 573.12.1.el6.x86_64
>>> >>>>>> KVM Version: 0.12.1.2 - 2.479.el6_7.2
>>> >>>>>> LIBVIRT Version: libvirt-0.10.2-54.el6_7.3
>>> >>>>>>
>>> >>>>>>
>>> >>>>>> On Wed, Jan 6, 2016 at 1:25 PM, Oved Ourfali <oourfali at redhat.com>
>>> >>>>>> wrote:
>>> >>>>>>>
>>> >>>>>>> All hosts (not VMs) in your cluster are centos 6.7?
>>> >>>>>>>
>>> >>>>>>> On Wed, Jan 6, 2016 at 9:52 AM, Budur Nagaraju
>>> >>>>>>> <nbudoor at gmail.com>
>>> >>>>>>> wrote:
>>> >>>>>>>>
>>> >>>>>>>> Os type is set to "other OS " and the centos 6.7 is installed
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>> On Wed, Jan 6, 2016 at 1:04 PM, Oved Ourfali
>>> >>>>>>>> <oourfali at redhat.com>
>>> >>>>>>>> wrote:
>>> >>>>>>>>>
>>> >>>>>>>>> Seems like you've mixed different versions in the same cluster,
>>> >>>>>>>>> which is the reason the host is non-operational.
>>> >>>>>>>>> What's the OS type and version of the hosts in your cluster?
>>> >>>>>>>>>
>>> >>>>>>>>> On Wed, Jan 6, 2016 at 7:59 AM, Budur Nagaraju
>>> >>>>>>>>> <nbudoor at gmail.com>
>>> >>>>>>>>> wrote:
>>> >>>>>>>>>>
>>> >>>>>>>>>>
>>> >>>>>>>>>>
>>> >>>>>>>>>> http://pastebin.com/jQiuHhZz
>>> >>>>>>>>>>
>>> >>>>>>>>>> On Wed, Jan 6, 2016 at 10:57 AM, Budur Nagaraju
>>> >>>>>>>>>> <nbudoor at gmail.com>
>>> >>>>>>>>>> wrote:
>>> >>>>>>>>>>>
>>> >>>>>>>>>>> HI
>>> >>>>>>>>>>>
>>> >>>>>>>>>>> Getting the below error while adding node to the OE which is
>>> >>>>>>>>>>> created from Hosted eghine.
>>> >>>>>>>>>>>
>>> >>>>>>>>>>>
>>> >>>>>>>>>>> 2016-01-06 10:50:49,404 INFO
>>> >>>>>>>>>>> [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo]
>>> >>>>>>>>>>> (DefaultQuartzScheduler_Worker-83) [7ce02003] Host
>>> >>>>>>>>>>> 3577e25f-e25e-425d-82ac-f6c3502a281e : infraesx2 is already
>>> >>>>>>>>>>> in
>>> >>>>>>>>>>> NonOperational status for reason
>>> >>>>>>>>>>> MIXING_RHEL_VERSIONS_IN_CLUSTER.
>>> >>>>>>>>>>> SetNonOperationalVds command is skipped.
>>> >>>>>>>>>>>
>>> >>>>>>>>>>>
>>> >>>>>>>>>>> Any way to fix the issue ? Engine version is 3.5.6.2-1.el6
>>> >>>>>>>>>>>
>>> >>>>>>>>>>> Thanks,
>>> >>>>>>>>>>> Nagaraju
>>> >>>>>>>>>>>
>>> >>>>>>>>>>
>>> >>>>>>>>>>
>>> >>>>>>>>>> _______________________________________________
>>> >>>>>>>>>> Users mailing list
>>> >>>>>>>>>> Users at ovirt.org
>>> >>>>>>>>>> http://lists.ovirt.org/mailman/listinfo/users
>>> >>>>>>>>>>
>>> >>>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>
>>> >>>>>>
>>> >>>>>
>>> >>>
>>> >>
>>> >> _______________________________________________
>>> >> Users mailing list
>>> >> Users at ovirt.org
>>> >> http://lists.ovirt.org/mailman/listinfo/users
>>> >>
>>> >
>>> >
>>> >
>>> > --
>>> > Didi
>>> > _______________________________________________
>>> > Users mailing list
>>> > Users at ovirt.org
>>> > http://lists.ovirt.org/mailman/listinfo/users
>>>
>>
>>
>
>



-- 
Didi

More information about the Users mailing list