Hi,

You can manually do the resize (Steps here are for FC) :
- Move the Storage Domain to maintenance
- First locate the LUN ID you want to refresh with :multipath -ll
On each hosts:
 - Run : rescan-scsi-bus.sh
 - Run: multipathd resize map 3600a0b800075de8f00000e3a52ddf25f   (replace
with your LUN_ID)
 - Check that the size was updated in multipat -ll
On the SPM:
 - Run: pvresize  /dev/mapper/3600a0b800075de8f00000e3a52ddf25f
 - Check with vgs

Move the domain to active again.
I am not sure that the engine will see the new size immediately. An engine
restart might be needed.

Regards,

Fred

On Wed, Jun 28, 2017 at 9:18 AM, Enrico Becchetti Gmail <
enrico.becche...@gmail.com> wrote:

>  Hi All,
> There are other ways to avoid using two lun ? If necessary I can also
> shutdown the entire cluster.
> thanks a lot !
> Bye
> Enrico
>
>
>
> Il 26/06/2017 12:32, Fred Rolland ha scritto:
>
> Hi,
>
> The support for refreshing the LUN size in an existing storage domain has
> been introduced in 3.6, so it is not relevant for you.
>
> You can add an additional LUN to an existing storage domain.
>
> Regards,
>
> Fred
>
> [1] https://www.ovirt.org/develop/release-management/features/
> storage/lun-resize/
>
> On Mon, Jun 26, 2017 at 9:35 AM, Enrico Becchetti Gmail <
> enrico.becche...@gmail.com> wrote:
>
>> Dear All,
>> I've oVirt 3.5.3-1-1.el6 with two kind of storage , NFS and Fibre
>> Channel.  Each Hypervisor
>> has HBA controller and Data CLutser has four Storage named: DATA_FC,
>> DATA_NFS, EXPORT
>> and ISO-DOMAIN.
>>
>> Anybody know how can I extend a DATA_FC ? Which is the best practies ?
>> Extend LUN from HP MSA controller or adding a new lun ?
>> if I extend lun inside HP MSA oVirt see the new size ?
>> Enrico
>> Thanks a lot.
>> Best Regards
>>
>> _______________________________________________
>> Users mailing list
>> Users@ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>
>
>
>
_______________________________________________
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Reply via email to