[Gluster-users] Shards and heal question post move-brick

[David Gossage]

On Thu, Jul 21, 2016 at 5:19 AM, David Gossage <dgossage at carouselchecks.com>
wrote:

>
>
> On Thu, Jul 21, 2016 at 4:54 AM, Lindsay Mathieson <
> lindsay.mathieson at gmail.com> wrote:
>
>> On 21/07/2016 7:10 PM, David Gossage wrote:
>>
>>> Biggest issue will be changing cache mode of disks which oVirt doesn't
>>> expose easily that you had mentioned would now be needed
>>>
>>
>> Just to clarify, I mean the caching mode of the qemu disks, not the
>> underlying host filesystem.
>>
>>
> I follow, and I think it an be set just not as a checkbox type flag.  I
> think you set via  custom flag per disk or vm.
> What I don't know is if any of the storage tests they do use flags that I
> wouldn't have access to that could fail.
> I've asked in ovirt users list about it and if anyone is using gluster
> 3.7.13 in their 3.6.x line.  Redhat gives the impression that due to their
> RHS gluster is supposed to be well integrated so I am hoping to get some
> feedback on this.
>

found it
oVirt -> edit vm settings -> custom properties ->  key viodiskcache :
none|writeback|writethrough



>
>> oVirt doesn't let you touch that? how annoying, its a useful tool and can
>> make a huge difference to performance.
>>
>> --
>> Lindsay Mathieson
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160721/fe32b593/attachment.html>